This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
50% To Boiling Acetone, To Boiling Aluminum Chloride, < 10%, 70~ Aluminum Chloride, > 10%, 70 oF. Aluminum Chloride, < 10%, To Boiling
X X X A X X X
B
Corrosion Resistant Alloys All Iron Bronze fitted or Standard fitted All Bronze Corrosion Resistant Alloy Steels
This chart is intended as a guide in the selection of economical materials. It must be kept in mind that corrosion rates may vary widely with temperature, concentration, and the presence of trace elements or abrasive solids. Blank spaces indicate a lack of accurate corrosion information for those specific conditions.
(5) Crevice or Concentration Cell Corrosion occurs in joints or small surface imperfections. Portions of the liquid become trapped and a difference in potential is established due to the oxygen concentration difference in these cells. The resulting corrosion may progress rapidly leaving the surrounding area unaffected.
Corrosive
14 10 8 6 4
Materials Selection Chart
(4) Pitting Corrosion is a localized rather than uniform type of attack. It is caused by a breakdown of the protective film and results in rapid pit formation at random locations on the surface.
steel C.I. D.I.
Material of Construction
Brz.
316SS
GA-20
CD4MCu
Mon
Ni
H-B
A
A
A A A A A B B C
A A B C A C C X
A B B B A B C X
A B B B A C X X
A C X A A A A
A B X A B X X
A A B A C C X
H-C A A A A A B B X
A A A A A A A A Cont.
Appendix 7A: Centrifugal Pump Fundamentals 261
Corrosive
steel C.I. D.I.
Brz.
316SS
GA-20
CD4MCu
Mon
Ni
H-B
H-C
TI
Zl
Aluminum Chloride, >10%, To Boiling Aluminum Sulphate, 70 ~ Aluminum Sulphate, < 10%, To Boiling Aluminum Sulphate, > 10%, To Boiling Ammonium Chloride, 70~ Ammonium Chloride, < 10%, To Boiling Ammonium Chloride, > 10%, To Boiling Ammonium Fluosilicate, 70 ~ Ammonium Sulphate, < 40%, To Boiling Arsenic Acid, to 225~ 9
X X X X X X X X X X
X B B C X X X X X X
X A B C B B X C B C
X A A B B B C B B B
X A B C B C X C C C
X B X X B B C X B X
X B X X B B C X B X
A B A B
X B A B A A C C B
X A A C A A C X A
A A A B A A C X A
Barium Chloride, 70~ < 30% Barium Chloride, < 5%,To Boiling Barium Chloride, > 5%, To Boiling Barium Hydroxide, 70 ~ Barium Nitrate, To Boiling Barium Sulphide, 70 ~ Benzoic Acid Boric Acid, To Boiling Boron Trichloride, 70 ~ Dry Boron Trifluoride, 70 ~F. 10%, Dry Brine (acid), 70~ Bromine (dry), 70~ Bromine (wet), 70~
X X X B C C X X B B X X X
B B C X X X C C B B X X X
C C X A B B B B B B X X X
B B C A B B B B B A X X X
C C X A B B B B B B X X X
B B C B
B B C B B
B B C B
X B C B A
B B C A B X B C B A
A A B
B A C A B A A B
X X
C C
A A B A B B B
B A C A B A A B B X X
X X
Calcium Bisulphite, 70 ~ Calcium Bisulphite, To Hot Calcium Chloride, 70~ Calcium Chloride, < 5%, To Boiling Calcium Chloride, > 5%, To Boiling Calcium Hydroxide, 70~ Calcium Hydroxide, < 30%, To Boiling Calcium Hydroxide, > 30%, To Boiling Calcium Hypochlorite, < 2%, 70~ Calcium Hypochlorite, > 2%, 70~ Carbolic Acid, 70~ (phenol) Carbon Bisulphide, 70 ~ Carbonic Acid, 70~ Carbon Tetrachloride, Dry to Boiling Chloric Acid, 70~ Chlorinated Water, 70~ Chloroacetic Acid, 70~ Chlorosulphonic Acid, 70 ~ Chromic Acid, < 30% Citric Acid Copper Nitrate, to 175 ~F. Copper Sulphate, To Boiling Cresylic Acid Cupric Chloride Cyanohydrin, 70~
X X B C X B C X X X C B B B X C X X X X X X C X C
X X C C C B B X X X B B C B X C
B C B B C B B C X X A A A A X B X X C A B C B X B
B B B B B B B C C C A A A A B B X C B A B B B X B
B C B B C B B C X X A A A A C B
X X B A C B B C X X A B C A X
X X B A C B B C X X A B B A X
X C A B C B X B
X X C X X C C
Dichloroethane Diethylene Glycol, 70~ Dinitrochlorobenzene, 70~
C A C
B B B
B A A
B A A
B A A
Ethanolamine, 70~ Ethers, 70~ Ethyl Alcohol, To Boiling Ethyl Cellulose, 70 ~ Ethyl Chloride, 70 ~ Ethyl Mercaptan, 70~ Ethyl Sulphate, 70~ Ethylene Chlorohydrin, 70 ~ Ethylene Dichloride, 70 ~ Ethylene Glycol, 70~ Ethylene Oxide, 70~
B B A A C C C C C B C
X B A B B X B B B B X
B B A B B B B B B B B
B A A B A A A B B B B
Ferric Chloride, < 5%, 70 ~ Ferric Chloride, > 5%, 70 ~ Ferric Nitrate, 70 ~ Ferric Sulphate, 70 ~ Ferrous Sulphate, 70 ~
X X X X X
X X X X C
X X B C C
X X A B B
(dry)
X X C X C C C
B X
B
A A A
A
B C A A A A A B A B A
A B X
A B C A
X X C X X C X
A
A B A X A B C
C B A
B B A
B B A
B B A
B A A B B B B B B B B
C B A B B
X B A B B
B B B B B
B A B B B
B A B B B
B B B B
B B A A
X X B C C
X X X C C
X X X C C
X X
A X B
B
A A A A B A A A A A A A A A
A A A A B
A B A A A
A A B A A B A
A B X A A
B
X
A A A
B A A
A A A A A
A A A A A
B C A A
A A A A
A A A A
A B B B B
A B
B X
B A
B A
A
Cont.
262 Process Plant Machinery steel C.I. D.I.
Brz.
316SS
GA-20
Formaldehyde, To Boiling Formic Acid, to 212~ Freon, 70~
B X A
B C A
A X A
A A A
B A A
A C A
A A A
Hydrochloric Acid, < 1%, 70 ~F. Hydrochloric Acid, 1 - 20%, 70 ~F. Hydrochloric Acid, > 20%, 70 ~ Hydrochloric Acid, < 1/2%, 175~ Hydrochloric Acid, 1/2-2%,175 ~ Hydrocyanic Acid, 70~ Hydrogen Peroxide, < 30% < 150 ~ Hydrofluoric Acid, < 20%, 70 ~ Hydrofluoric Acid, > 20%, 50 ~ Hydrofluoric Acid, To Boiling Hydrofluorsilicic Acid, 70~
X X X X X X C X X X X
X X X X X X X B C X
C X X C X C B X X X C
B X X C
A C C C C C B B B C B
B X X X X
A A B A A
A X X X
A X X X
Lactic Acid, < Lactic Acid, > Lactic Acid, < Lime Slurries,
X X X B
B B X B
A B C B
A B B B
X C X B
C C X B
B B B B
B B B B
A A A B
A A A B
Magnesium Chloride, 70 ~ Magnesium Chloride, < 5%, To Boiling Magnesium Chloride, > 5%, To Boiling Magnesium Hydroxide, 70 ~ Magnesium Sulphate Maleic Acid Mercaptans Mercuric Chloride, < 2%, 70~ Mercurous Nitrate, 70 ~ Methyl Alcohol, 70 ~
C X X B C C A X C A
C C C A C C X X X A
B C X B B B A X B A
A B C B A B A X B A
C C C B B C X X C A
C C C A B C X C
A A B B C B
A A B B C B
A A B A B A
A A B
A
A
A
A
B C A
A
A
Naphthalene Sulphonic Acid, 70 ~ Naphthalenic Acid, To Hot Nickel Chloride, 70 ~ Nickel Sulphate Nitric Acid Nitrobenzene, 70~ Nitroethane, 70~ Nitropropane, 70 ~ Nitrous Acid, 70~ Nitrous Oxide, 70~
X C X X X A A A X C
C C X C X C A A X C
B B C B B A A A X C
B B B B B A A A C C
C C C C X B A A X X
C C X C X B A A X X
B B A
B B
B
B A A
B A A
B A B
Oleic Acid Oleum, 70~ Oxalic Acid
C B X
C X C
B B C
B B B
C X C
C X C
C B B
C B B
Palmitic Acid Phenol (see carbolic acid) Phosgene, 70~ Phosphoric Acid, < 10%, 70 ~F. Phosphoric Acid, > 1 0 - 70%, 70 ~F. Phosphoric Acid, < 20%, 175 ~ Phosphoric Acid, > 20%, 175 ~ < 85% Phosphoric Acid, > 10%, Boil, < 85% Phthalic Acid, 70~ Phthalic Anhydride, 70~ Picric Acid, 70~ Potassium Carbonate Potassium Chlorate Potassium Chloride, 70 ~ Potassium Cyanide, 70 ~ Potassium Dichromate Potassium Ferricyanide Potassium Ferrocyanide, 70 ~ Potassium Hydroxide, 70 ~ Potassium Hypochlorite Potassium Iodide, 70 ~ Potassium Permanganate Potassium Phosphate
B
B
B
A
B
B
C X X X X X C B X B B C B B C X C X C B C
C C C C C C B C X B C C X B B B C C B B C
B A A B C X B A C A A B B A B B B C B B B
B A A B B C A A B A A A B A B B A B B B B
C C C C C C B A C B C B C B B B A X B C
C C C C C C B A X B C B C B B B A X B B
B A B A B C B A
B A C A C C B A B B B B B B B B C B B B
Sea Water, 70~ Sodium Bisulphate, 70 ~
C X
B C
B C
A B
Corrosive
50%, 70 ~F. 50%, 70 ~F. 5%, To Boiling 70 ~
CD4MCu
H-B
Mon
B B B C X B
H-C
B
B A A A
B
A A
C
B B B B B B B
A B
C B X
C
A B C C C A
A B B C C A
A A A
A A A
A A
A A B A
B A A
A
A
B
B
A B
A A Cont.
Appendix 7A: Centrifugal Pump Fundamentals 263 steel C.I. D.I.
Brz.
316SS
GA-20
CD4MCu
Mon
Ni
H-B
H-C
Sodium Bromide, 70~ Sodium Carbonate Sodium Chloride, 70~ Sodium Cyanide Sodium Dichromate Sodium Ethylate Sodium Fluoride Sodium Hydroxide, 70 ~ Sodium Hypochlorite Sodium Lactate, 70 ~ Stannic Chloride, < 5%, 70~ Stannic Chloride, > 5%, 70 ~ Sulphite Liquors, To 175 ~ Sulphure (molten) Sulphur Dioxide (spray), 70 ~ Sulphuric Acid, < 2%, 70~ Sulphuric Acid, 2-40%, 70~ Sulphuric Acid, 40%, < 90%, 70 ~ Sulphuric Acid, 93-98%, 70~ Sulphuric Acid, < 10%, 175~ Sulphuric Acid, 10-60% & > 80%, 175~ Sulphuric Acid, 60-80%, 175~ Sulphuric Acid, < 3/4%, Boiling Sulphuric Acid, 3/4-40%, Boiling Sulphuric Acid, 4 0 - 6 5 % & > 85%, Boil Sulphuric Acid, 65-85%, Boiling Sulphurous Acid, 70 ~F.
B B C B B B C B X B X X X B C X X X B X X X X X X X X
C B B X X A C B X C C X C X C C C X X C X X X X X X C
B B B B B A B B C C X X B A B B C X B X X X C X X X C
B A B B B A B A C C C X B A B A B B B B B X B C X X B
B B B B B A B B C C X X B A B B C X B X X X C X X X C
B B A X
B B A X
B B B
B B B
A B A X C C X C C C C C X X X X X X X X X X
A B A X
C A
C X C C C C C X X X X X X X X X X
A A A B A B B B B X X B
C A B C B C B A B A A A B C C C B C X X B
Titanium Tetrachloride, 70 ~ Trichlorethylene, To Boiling
C B
C
C B
B B
C B
C B
B
B
Urea, 70~
C
C
B
B
B
C
C
Vinyl Acetate Vinyl Chloride
B B
B C
B B
B B
B B
C
Water, To Boiling
B
A
A
A
A
A
Zinc Chloride Zinc Cyanide, 70~ Zinc Sulphate
C X X
C B C
B B A
A B A
B B A
B B C
Corrosive
Piping Design The design of a piping system can have an important effect on the successful operation of a centrifugal pump. Such items as sump design, suction piping design, suction and discharge pipe size, and pipe supports must all be carefully considered. Selection of the discharge pipe size is primarily a matter of economics. The cost of the various pipe sizes must be compared to the pump size and power cost required to overcome the resulting friction head.
TI
Zl
A A B B
A A
B A A
B A B
A B A A C B X X X X X X X X X X A
A B
A A C C B C C B B X X B
C B
A
A
C
C
B
B
C
C
B B
A
A
A
A
A
A
B B C
B B C
B C
A B A
A B
C B B C
separation keeps the liquid from evenly filling the impeller. This upsets hydraulic balance leading to vibration, possible cavitation, and excessive shaft deflection. Shaft breakage or premature bearing failure may result. On pump installations involving suction lift, air pockets in the suction line can be a source of trouble. The suction pipe should be exactly horizontal, or with a uniform slope upward from the sump to the pump as shown in Figure 7A-17. There should be no high spots where air can collect and cause the pump to lose its prime. Eccentric rather than concentric reducers should always be used.
The suction pipe should never be smaller than the suction connection of the pump, and in most cases should be at least one size larger. Suction pipes should be as short and as straight as possible. Suction pipe velocities should be in the 5 to 8 feet per second range unless suction conditions are unusually good.
If an elbow is required at the suction of a double suction pump, it should be in a vertical position if at all possible. Where it is necessary for some reason to use a horizontal elbow, it should be a long radius elbow and there should be a minimum of two diameters of straight pipe between the elbow and the pump as shown in Figure 7A-18. Figure 7A-19 shows the effect of an elbow directly on the suction. The liquid will flow toward the outside of the elbow and result in an uneven flow distribution into the two inlets of the double suction impeller. Noise and excessive axial thrust will result.
Higher velocities will increase the friction loss and can result in troublesome air or vapor separation. This is further complicated when elbows or tees are located adjacent to the pump suction nozzle, in that uneven flow patterns or vapor
There are several important considerations in the design of a suction supply tank or sump. It is imperative that the amount of turbulence and entrained air be kept to a minimum. Entrained air will cause reduced capacity and efficiency as
The suction piping size and design is far more important. Many centrifugal pump troubles are caused by poor suction conditions.
264 Process Plant Machinery
CHECK VALVE
ECCENTRIC REDUCER
GATE VALVE
LONG RADIUS
(la) CORRECT
FOOT VALVE (IF USED)
CHECK VALVE
ECCENTRIC REDUCER
LONG RADIUS ELBOW
f
GATE VALVE
SUCTION PIPE SLOPES UPWARDS FROM SOURCE OF SUPPLY
(lb) CORRECT FOOT VALVE (IF USED) STRAINER AIR POCKET BECAUSE ECCENTRIC REDUCER IS NOT USED AND BECAUSE SUCTION PIPE DOES NOT SLOPE GRADUALLY UPWARD FROM SUPPLY
GATE VALVE
/ CHECK GATE VALVE SHOULD NOT BE BETWEEN VALVE CHECK VALVE AND PUMP
(1r WRONG
Fig. 7A-17
Air pockets in suction piping
well as vibration, noise, shaft breakage, loss of prime, and/or accelerated corrosion. The free discharge of liquid above the surface of the supply tank at or near the pump suction can cause entrained air to enter the pump. All lines should be submerged in the tank, and baffles should be used in extreme cases as shown in Figure 7A-20. Improper submergence of the pump suction line can cause a vortex which is a swirling funnel of air from the surface directly
into the pump suction pipe. In addition to submergence, the location of the pipe in the sump and the actual dimensions of the sump are also important in preventing vortexing and/or excess turbulence. For horizontal pumps, Figure 7A-21 can be used as a guide for minimum submergence and sump dimensions for flows up to approximately 3000 gpm. Baffles can be used to help prevent vortexing in cases where it is impractical or impossible to maintain the required submergence. Figure 7A-22 shows three such baffling arrangements.
Appendix 7A" Centrifugal Pump Fundamentals 265
Fig. 7A-18 Elbows at pump suction
Fig. 7A-19 Effect of elbow directly on suction
Fig. 7A-20 Keeping air out of pump
266 Process Plant Machinery
Flg. 7A-21
Mlnlmum suctlon plpe submergence and sump dlmenslons
Fig. 7A-22
Baffle arrangements for vortex prevention
Large units (over 3000 gpm) taking their suction supply from sumps, especially vertical submerged pumps, require special attention. The larger the unit, the more important the sump design becomes.
be located near the back wall and should not be subjected to rapid changes in direction of the flow pattern. The velocity of the water in the area of the suction pipes should be kept below one foot per second to avoid air being drawn into the pump.
Figure 7A-23 illustrates several preferred piping arrangements within a multiple pump pit. Note that the pipe should always
On horizontal pumps, a bell should be used on the end of the suction pipe to limit the entrance velocity to 3.5 feet
Appendix 7A: Centrifugal Pump Fundamentals 267 RECOMMENDED
NOT RECOMMENDED ~
Q
LT
0~
V, ,-, 1 fps OR LESS
V, = 2 tpS & UP
~
= t 'h TO 20
.-=
i
i
9
IF A = LESS THAN
A
0 o
NESS TO (~ DIST. .oo ROUND OR OGIVE
G
0
mmmm
WALL ENDS. GAP AT REAR OF WALL APPX. O13
@
d
,,mmm
0
~
i
0
i
i
9/16 D
-- A
PREFERED oc = 75"
Fig. 7A-23
Piping arrangements within multiple pump pits
per second. Also, a reducer at the pump suction flange to smoothly accelerate and stabilize the flow into the pump is desirable. The submergence of the suction pipe must also be carefully considered. The amount of submergence required depends upon the size and capacity of the individual pumps as well as on the sump design. Past experience is the best guide for determining the submergence. The pump manufacturer should be consulted for recommendations in the absence of other reliable data.
Stuffing B o x S e a l i n g The stuffing box of a pump provides an area in which to seal against leakage out of the pump along the shaft. Packing and mechanical seals are the two devices used to accomplish this seal.
Packing A typical packed stuffing box arrangement is shown in Figure 7A-24. It consists of: A)Five rings of packing. B)A lantern ring used for the injection of a lubricating and/or flushing liquid, and C ) A gland to hold the packing and maintain the desired compression for a proper seal. The function of packing is to control leakage and not to eliminate it completely. The packing must be lubricated, and a flow of from 40 to 60 drops per minute out of the stuffing box must be maintained for proper lubrication.
Lantern Sealing Liquid PackingGland Stufting Ring~ (Quench - ~ , ~ ,,~ Connection / Type) Box ~ i II \ ~ / Bushing ~ "~J" t , !1a ~. ~ . q 7 ~I ,....L (
J
i~\'3 . . . . . . . . . . . .
- (~
H
kZ
U
S'oUx
Throat
Fig. 7A-24 of parts)
-
~ ---~
"
Mechanical Packing
Typical stuffing box arrangement (description
The method of lubricating the packing depends on the nature of the liquid being pumped as well as on the pressure in the stuffing box. When the pump stuffing box pressure is above atmospheric pressure and the liquid is clean and nonabrasive, the pumped liquid itself will lubricate the packing Figure 7A-25. When the stuffing box pressure is below atmospheric pressure, a lantern ring is employed and lubrication is injected into the stuffing box (Figure 7A-26). A bypass line from the pump discharge to the lantern ring connection is normally used providing the pumped liquid is clean.
268 Process Plant Machinery
Lantern Ring Location F
Injected Fluid From
Thick SlurJ Including Paper Stock
Positive Fluid Pressure
Above Atmospheri, Pressure
eric re
Leakage Into _ Pump
Fig. 7A-25 Typical stuffing box arrangement when stuffIng box pressure is above atmosphere pressure
Fig. 7A-27 Typical stuffing box arrangement when pump ing slurries
Mechanical Seals The Basic Seal ,tic 3
Leakage Into _ Pump
A mechanical seal is a sealing device which forms a running seal between rotating and stationary parts. The design of liquid handling equipment with rotating parts today would include the consideration for the use of mechanical seals. Advantages over conventional packing are as follows: 1. Reduced friction and power losses. 2. Zero or limited leakage of product. 3. Elimination of shaft or sleeve wear. 4. Reduced maintenance. 5. Ability to seal higher pressures and more corrosive environments.
Fig. 7A-26 Typical stuffing box arrangement when stuffIng box pressure is below atmospheric pressure
The wide variety of styles and designs together with extensive experience allows the use of seals on most pump applications. A mechanical seal must seal at three points:
When pumping slurries or abrasive liquids, it is necessary to inject a clean lubricating liquid from an external source into the lantern ring (Figure 7A-27). A flow of from .2 to .5 gpm is desirable and a valve and flowmeter should be used for accurate control. The seal water pressure should be from 10 to 15 psi above the stuffing box pressure, and anything above this will only add to packing wear. The lantern ring is normally located in the center of the stuffing box. However, for extremely thick slurries like paper stock, it is recommended that the lantern ring be located at the stuffing box throat to prevent stock from contaminating the packing. The gland shown in Figures 7A-24 to 27 is a quench type gland. Water, oil, or other fluids can be injected into the gland to remove heat from the shaft, thus limiting heat transfer to the bearing frame. This permits the operating temperature of the pump to be higher than the limits of the bearing and lubricant design. The same quench gland can be used to prevent the escape of a toxic or volatile liquid into the air around the pump. This is called a smothering gland, with an external liquid simply flushing away the undesirable leakage to a sewer or waste receiver.
1. Static seal between the stationary part and the housing. 2. Static seal between the rotary part and the shaft. 3. Dynamic seal between the rotating seal face and the stationary seal face. Figure 7A-28 shows a basic seal with these components: 1. Stationary seal part positioned in the housing with preload on the "O" ring to effect sealing and prevent rotation. 2. Rotating seal part positioned on the shaft by the "O" ring. The "O" ring seals between it and the shaft and provides resiliency. 3. The mating faces. The faces are precision lapped for a flatness of 3 light bands and a surface finish of 5 microinches. 4. Spring assembly, rotates with the shaft and provides pressure to keep the mating faces together during periods of shut down or lack of hydraulic pressure.
Appendix 7A: Centrifugal Pump Fundamentals 269
HOUSING
_
X
/
\
_
SPRING ASSEMBLY
STATIONARY SEAL \ ROTATING SEAL PART PART / MATING FACES
Fig. 7A-28
INJECTED FLUID FROM PUMP VOLUTE /
Basic mechanical seal
FLUSH GLAND
PLAIN
~ \ ~ l . r ~
CLAMPED ,N
INJECTED FLUID .o...,.o
,J.;,~'"%mi;Ei~iS')
Flg. 7A-29
Slngle, Inslde, unbalanced seal
Flg. 7A-31
CIRCULATED FLUID INLET CONNECTION
RESTRICTING INJECTED BUSHING FLUID PLAIN
~\~-~y,~
~
~ ...~
.
.
Fig. 7A-30
.
.
.
.
.
.
.
.
.
.
CLAMPED IN
IONARY SEAT .
.
.
.
.
Slngle, Inslde, balanced seal
OUTLET CONNECTION FLUSH GLAND
~._
Single, outside, unbalanced seal
Fig. 7A-32
Double, inside, unbalanced seal
Fig. 7A-33
Tandem, Inside unbalanced seals
5. Driving member, positions the spring assembly and the rotating face. It also provides the positive drive between shaft and the other rotating parts. As wear takes place between the mating faces, the rotating face must move along the shaft to maintain contact with stationary face. The "O" ring must be free to move. A mechanical seal operates as each basic component performs its duty. Liquid pressure in the seal chamber forces the faces together and provides a thin film of lubricant between them. The faces, selected for low frictional qualities, are the only rubbing parts. These basic components are a part of every seal. The form, shape, style and design will vary greatly depending on service and manufacture. The basic theory, however, remains the same.
270 Process Plant Machinery Types Mechanical seals can be classified into the general types and arrangements shown below. Understanding these classes provides the first step in proper seal selection. (a) Single seals-Inside, outside, unbalanced, balanced (b) Double seals- Unbalanced or balanced Single Seal, Inside Unbalanced
The single inside seal mounts on the shaft or sleeve within the stuffing box housing. The pumpage is in direct contact with all parts of the seal and provides the lubrication for the faces. The full force of pressure in the box acts on the faces providing good sealing to approximately 100 P.S.I.G. This is the most widely used type for services handling clear liquids. A circulation or by-pass line connected from the volute to the stuffing box provides continual flushing of the seal chamber. Single Seals, Outside Unbalanced
This type mounts with the rotary part outside of the stuffing box. The springs and drive element are not in contact with the pumpage, thus reducing corrosion problems and preventing product accumulation in the springs. Pressures are limited to the spring rating, usually 35 P.S.I.G. Usually the same style seal can be mounted inside or outside. The outside seal is easier to install, adjust and maintain. A restricting bushing can be used to control leakage of an external sealing liquid into the pumpage. Single Seals, Balanced Balancing a seal varies the face loading exerted by the box pressure, thus extending the pressure limits of the seal. A balanced rotating part utilizes a stepped face and a sleeve. Balanced seals are used to pressures of 2000 P.S.I.G. Their use is also extensive on light hydrocarbons which tend to vaporize easily. Balanced outside seals allow box pressure to be exerted toward the seal faces, thus allowing pressure ranges to above 150 P.S.I. as compared to the 35 P.S.I.G. limit for the unbalanced outside seal. Double Seals
Double seals use two seals mounted back to in the stuffing box. The stuffing box is pressurized with a clear liquid from an external source. This liquid is circulated thru the double seal chamber at 1/4-1 GPM to cool and lubricate the mechanical
seals. Double seals are used on solutions that contain solids, are toxic or extremely corrosive. The external source fluid should be compatible with the pumpage. TANDEM SEALS
A variation of the double seal arrangement. The purpose of this seal is to provide a backup seal in the event the primary seal fails. The primary mechanical seal functions in a manner identical to that of a conventional single inside seal. The cavity between the primary seal and the backup seal is flooded with liquid to provide lubrication for the backup seal. The seal arrangement is used on toxic or hazardous chemical, and transfer and pipeline services to provide an extra measure of safety and allow equipment to operate until time to shut down. Selection The proper selection of a mechanical seal can be made only if the full operating conditions are known. These conditions are as follows: 1. Liquid 2. Pressure 3. Temperature 4. Characteristics of Liquid 1. Liquid Identification of the exact liquid to be handled provides the first step in seal selection. The metal parts must be corrosion resistant. These pads, usually available in steel, bronze, stainless steel, or Hastelloy, provide a wide choice to meet specific services. The mating faces must also resist both corrosion and wear Carbon, ceramic glass-filled Teflon, Stellite or tungsten carbide are available and offer both excellent wear properties and corrosion resistance. Stationary sealing members of synthetic rubber, asbestos and Teflon complete the proper material selection. 2. Pressure The proper type of seal, unbalanced or balanced, is based on the pressure on the seal and on the seal size. Figure 7A-34 shows the normal limits for unbalanced seals of various types. 3. Temperatures The temperature will in part determine the use of the sealing members Synthetic rubbers are used to approximately 400 F., Teflon to 500 F. and asbestos to
:I k
I
Maximum stuffing box pressures for Unbalanced Mechanical Seals on water solutions 160" max. Unbalanced seals generally limited to 200 PSIG maximum stuffing box pressure. All ratings
)400 =300 L
}=
i ~0o =,
m
.
\
9 -'~ ~
\
.~
~~S~.o r ~
~
based on one carbon face against hard face:
I
'
~
_
f .
I -CERAMIC 2-NI-RESIST ~
_
3-$TELI.ITE (not generally reeolIMtended on water sendr
4-TUNGSTEN CARBIDE
1 2 3 4 5 6 SEAL SIZE--INCHES (8Mfl or Sleeve O.D.)
Fig. 7A-34
Pressure-velocity limits, unbalanced seals
Appendix 7A: Centrifugal Pump Fundamentals 271 750 F. Cooling the liquid in the seal chamber by water cooling jackets or cool liquid flushing, often extends seal life and allows wider selection of materials.
L...
._i
,p_.._
4. Characteristics of Liquid Abrasive liquids create excessive wear and short seal life. Double seals or clear liquid flushing from an external source allows the use of mechanical seals on these difficult liquids. On light hydrocarbons balanced seals are often used to promote longer seal life, even though pressures are low.
II
Environmental Controls
Flg. 7A-36
Environmental controls are necessary for reliable performance of a mechanical seal on many applications. Pump manufactures and the seal vendors offer a variety of arrangements to combat these problems.
services, heat is provided to the jacket to melt or prevent the liquid from freezing (liquid sulfur).
1. Corrosion
DIRTY or INCOMPATIBLE ENVIRONMENTS
2. Temperature Control 3. Dirty or incompatible environments
CORROSION Corrosion can be controlled by selecting seal materials that are not attacked by the pumpage. When this is difficult, external fluid injection of a non-corrosive chemical to lubricate the seal is possible. Single or double seals could be used, depending on if the customer can stand delusion of his product.
TEMPERATURE CONTROL As the seal rotates, the faces are in contact. This generates heat and if this heat is not removed, the temperature in the stuffing box can increase and cause sealing problems. A simple by-pass flush of the product over the seal faces will remove the heat generated by the seal (Figure 7A-35). For higher temperature services, by-pass of product through a cooler may be required to cool the seal sufficiently (Figure 7A-36). External cooling fluid injection can also be used. Jacketed stuffing boxes are used on many pumps to cool the environment around the mechanical seal (Figure 7A-29). This will also allow the use of a mechanical seal on services where it would not normally function (hot heat transfer oil). For other
Mechanical seals do not normally function well on liquids which contain solids or can solidify on contact with the atmosphere. Here, by-pass flush through a filter, a cyclone separator or a strainer are methods of providing a clean fluid to lubricate the stuffing box. Strainers are effective for particles larger than the openings on a 40 mesh screen. Cyclone separators are effective on solids 10 micron or more in diameter, if they have a specific gravity of 2.7 and the pump develops a differential pressure of 30-40 psi. Filters are available to remove solids 2 micron and larger. If external flush with clean liquid is available, this is the most fail proof system. Lip seal or restricting bushings are available to control flow of injected fluid to flows as low as 1/2 GPM (Figure 7A-30). Quench type glands are used on fluids which tend to crystalize on exposure to air. Water or steam is put through this gland to wash away any build up. Other systems are available as required by the service.
Field Testing Methods A. Determination of total head The total head of a pump can be determined by gauge readings as illustrated in Figure 7A-37.
o
h
=
ha
_
Flg. 7A-35
Fig. 7A-37 readings
Determination
of total
head from
gauge
272
Process Plant Machinery
Negative Suction Pressure: TDH = Discharge gauge reading converted to feet of liquid + vacuum gauge reading converted to feet of liquid + distance between point of attachment of vacuum gauge and the centerline of the discharge gauges, h, in feet +
( Vd2 Vs2 I 2g 2g
MERCURY
Positive Suction Pressure: or TDH = Discharge gauge reading converted to feet of liquid-pressure gauge reading in suction line converted to ft. of liquid + distance between center of discharge and suction gauges, h, in feet+ Vs2 2g 2g
( V~
)
In using gauges when the pressure is positive or above atmospheric pressure, any air in the gauge line should be vented off by loosening the gauge until liquid appears. This assures that the entire gauge line is filled with liquid and thus the gauge will read the pressure at the elevation of the centerline of the gauge. However, the gauge line will be empty of liquid when measuring vacuum and the gauge will read the vacuum at the elevation of the point of attachment of the gauge line to the pipe line. These assumptions are reflected in the above definitions. The final term in the above definitions accounts for a difference in size between the suction and discharge lines. The discharge line is normally smaller than the suction line and thus the discharge velocity is higher. A higher velocity results in a lower pressure since the sum of the pressure head and velocity head in any flowing liquid remains constant. Thus, when the suction and discharge line sizes at the gauge attachment points are different, the resulting difference in velocity head must be included in the total head calculation. Manometers can also be used to measure pressure. The liquid used in a manometer is normally water or mercury, but any liquid of known specific gravity can be used. Manometers are extremely accurate for determining low pressures or vacuums and no calibration is needed. They are also easily fabricated in the field to suit any particular application. Figure 7A-38 and 39 illustrate typical manometer set ups.
B. Measurement of capacity
Flg. 7A-39
Manometer Indlcatlng pressure
is often practical when pumping into an accurately measured reservoir or tank, or when it is possible to use small containers which can be accurately weighed. These methods, however, are normally suited only to relatively small capacity systems.
c.) Venturi meter A venturi meter consists of a converging section, a short constricting throat section and then a diverging section. The object is to accelerate the fluid and temporarily lower its static pressure. The flow is then a function of the pressure differential between the full diameter line and the throat. Figure 7A-40 shows the general shape and flow equation. The meter coefficient is determined by actual calibration by the manufacturer and when properly installed the Venturi meter is accurate to within plus or minus 1%.
d.) Nozzle A nozzle is simply the converging portion of a venturi tube with the liquid exiting to the atmosphere. Therefore, the same formula can be used with the differential head equal to the gauge reading ahead of the nozzle. Figure 7A-41 lists theoretical nozzle discharge flows.
e.) Orlflce An orifice is a thin plate containing an opening of specific shape and dimensions. The plate is installed in a pipe and the flow is a function of the pressure upstream of the orifice. There are numerous types of orifices available and their descriptions and applications are covered in the Hydraulic Institute Standards and the ASME Fluid Meters Report. Orifices are not recommended for permanent installations due to the inherent high head loss across the plate.
a.) Magnetic Flow Meter
f.) Weir
A calibrated magnetic flow meter is an accurate means of measuring flow in a pumping system. However, due to the expense involved, magnetic flow meters are only practical in small factory test loops and in certain process pumping systems where flow is critical.
A weir is particularly well suited to measuring flows in open conduits and can be adapted to extremely large capacity systems. For best accuracy, a weir should be calibrated in place. However, when this is impractical, there are formulas which can be used for the various weir configurations. The
b.) Volumetric measurement Pump capacity can be determined by weighing the liquid pumped or measuring its volume in a calibrated vessel. This
TI hi
~/ H Q(GPM) = S.67 CDz= 1 - R4 C = Instrument Coefficient DI = Entrance Diameter in Inches D2 -= Throat Diameter in Inches II R -- D21D; |I~ H = Differential Head in Inches = nl -- nzL "l"-
L ~ ~ l ~ Flg. 7A-38 vacuum
Manometer Indlcstlng
Flg. 7A-40
Venturl meter
Appendix 7A: Centrifugal Pump Fundamentals 273
Theoretical Discharge of Nozzles in U.S. GPM Head
Veloc'y of , Oisch. Feet Feet per Sec.
Lbs. 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150 175 200 250 300
I
i
23.1 38.6 34.6 47.25 46.2 54.55 57.7 61.0 69.3 66.85 80.8 72.2 92.4 77.2 103.9 81.8 115.5 86.25 127.0 90.4 138.6 94.5 150.1 98.3 161.7:102.1 , 173.2 i 105.7 1184.8 i 109.1 196.3 112.5 207.9 115.8 219.4 119.0 230.9 122.0 242.4 125.0 254.0 128.0 265.5 130.9 2 7 7 . 1 133.7 288.6 136.4 300.2 139.1 311,7 141.8 323.3 144.3 334.8 146.9 346.4 149.5 4 0 4 . 1 161.4 461.9 172.6 577.4 193.0 692.8 211.2
Diameter of Nozzle in Inches s/,6 I
0.37 0.45 0.52 0.58 0.64 0.69 0.74 0.78 0.83 0.87 0.90 0.94 0.98 1.01 1.05 1.08 1.11 1.14 1.17 1.20 1.23 1.25 1.28 1.31 1.33 1.36 1.38 1.41 1.43 1.55 1.65 1.85 2.02 1 s/z
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150 175 i 200 i 250 i 300 ,
23.1 34.6 46.2 57.7 69.3 80.8 92.4 103.9 115.5 127.0 138.6 150.1 161.7, 173.2 184.8 196.3 207.9 219.4 230.9 242.4 254.0 265.5 277.1 288.6 300.2 311.7 323.3 334.8 346.4 404.1 461.9 577.4 692.8
38.6 47.25 54.55 61.0 66.85 72.2 77.2 81.8 86.25 90.4 94.5 98.3 102.1 105.7 109.1 112.5 115.8 119.0 122.0 125.0 128.0 130.9 13;~.7 136.4 139.1 141.8 144.3 146.9 149.5 161.4 172.6 193.0 211.2
213 260 301 336 368 398 425 451 475 498 521 542 563 582 602 620 638 656 672 689 705 720 736 751 767 780 795 809 824 890 950 1063 1163
I
s/= 1.48 1.81 2.09 2.34 2.56 2.77 2.96 3.13 3.30 3.46 3.62 3.77 3.91 4.05 4.18 4.31 4.43 4.56 4.67 4.79 4.90 5.01 5.12 5.22 5.33 5.43 5.53 5.62 5.72 6.18 6.61 7.39 8.08
$/,6 I
3.32 4.06 4.69 5.25 5.75 6.21 6.64 7.03 7.41 7.77 8.12 8.45 8.78 9.08 9.39 9.67 9.95 10.2 10.5 10.8 11.0 11.2 11.5 11.7 12.0 12.2 12.4 12.6 12.9 13.9 14.8 16.6 18.2
1% 289 354 409 458 501 541 578 613 647 678 708 737 765 792 818 844 868 892 915 937 960 980 1002 1022 1043 1063 1082 1100 1120 1210 1294 1447 1582 ,
2 378 463 535 598 655 708 756 801 845 886 926 964 1001 1037 1070 1103 1136 1168 1196 1226 1255 1282 1310 1338 1365 1390 1415 1440 1466 1582 1691 1891 2070
~,~ !
5.91 7.24 8.35 9.34 10.2 11.1 11.8 12.5 13.2 13.8 14.5 15.1 15.7 16.2 16.7 17.3 17.7 18.2 18.7 19.2 19.6 20.0 20.5 20.9 21.3 21.7 22.1 22.5 22.9 24.7 26.4 29.6 32.4
; 2%
~ I
13.3 16.3 18.8 21.0 23.0 24.8 26.6 28.2 29.7 31.1 32.5 i 33.8
135.2
! 36.4 ; 37.6 38.8 39.9 41.0 42.1 43.1 44.1 45.1 46.0 47.0 48.0 48.9 149.8 I 50.6 51.5 55.6 59.5 66.5 72.8 2Vz
479 591 585 723 676 835 756 934 828 1023 895 1106 957 1182 1015 1252 1070 1320 1121 , 1385 1172 1447 1220 1506 1267 1565 1310 1619 1 3 5 4 1672 1395 1723 1436 1773 1476 1824 1512 1870 1550 1916 1588 1961 1 6 2 1 2005 1 6 5 9 2050 1690 2090 1726 2132 1759 2173 1790 2212 1820 2250 1853 2290 2000 2473 2140 2645 2392 2955 2615 3235
]
s/z I
23.6 28.9 33.4 37.3 40.9 44.2 47.3 50.1 52.8 55.3 57.8 60.2 62.5 64.7 66.8 68.9 70.8 72.8 74.7 176.5 78.4 80.1 81.8 83.5 85.2 86.7 88.4 89.9 91.5 98.8 106. 118. 129.
%
~
I
7/,
I
36.9 53.1 45.2 65.0 52.2 75.1 58.3 84.0 63.9 92.0 69.0 99.5 73.8 106. 78.2 113. 82.5 119. 86.4 125. 90.4 130. 94.0 136. 97.7 141. ;101. 146. !104. 150. 108. 155. 111. 160. 114. 164. 117. 168. 120. 172. 122. 176. : 125. 180. 128. 184 130. 188. 1133. 192. I 136. 195. 138. 199. 140. 202. 143. 206. 154. 222. 165. 238. 185. 266. 202. I 291.
2% 714 874 1009 1128 1236 1335 1428 1512 1595 1671 1748 1819 1888 1955 2020 2080 2140 2200 2255 2312 2366 2420 2470 2520 2575 2620 2670 2715 2760 2985 3190 3570 3900 ,
3
I
3s/z
851 1041 1203 1345 1473 1591 1701 1802 1900 1991 2085 2165 2250 2330 2405 2480 2550 2625 2690 2755 2820 2885 2945 3005 3070 3125 3180 3235 3295 3560 3800 4250 4650 ,
I
72.4 88.5 i 102. 114. 125. 135. 145. 153. 162. 169. 177. 184. 191. 198. 205. 211. 217. 223. 229. 234. 240. 245. 251. 256. 261. 266. i 271. 275. 280. 302. 323. 362. 396 4
1158 1418 1638 1830 2005 2168 2315 2455 2590 2710 2835 2950 3065 3170 3280 3375 3475 3570 3660 3750 3840 3930 4015 4090 4175 4250 4330 4410 4485 4840 5175 5795 6330
~i
1 94.5 116. 134. 149. 164. 177. 189. 200. 211. 221. 231. 241. 250. 259. 267. 276. 284. 292. 299. 306. 314. 320. 327. 334. 341. 347. 354. 360. 366. 395. 423. 473. 517. 4s/z
: 1510 1915 1850 2345 2135 2710 2385 3025 2615 3315 2825 3580 3020 3830 3200 4055 3375 4275 3540 4480 3700 4685 3850 4875 4000 5060 4135 5240 4270 5410 4400 5575 4530 5740 4655 5900 4775 6050 4890 6200 5010 6350 5120 6490 5225 6630 5340 6760 5450 6900 5550 7030 5650 7160 5740 7280 5850 7410 6310 8000 6750 8550 7550 9570 ~ 8260 , 10480
1% I
120 147 169 189 207 224 239 253 267 280 293 305 317 327 338 349 359 369 378 388 397 406 414 423 432 439 448 455 463 500 535 598 655 5 2365 2890 3340 3730 4090 4415 4725 5000 5280 5530 5790 6020 6250 6475 6690 6890 7090 7290 7470 7650 7840 8010 8180 8350 8530 8680 8850 8990 91 50 9890 10580 11820 12940
1 s~ I
148 181 209 234 256 277 296 313 330 346 382 376 391 404 418 431 443 456 467 479 490 501 512 522 533 543 553 562 572 618 660 739 808 51/z 2855 3490 4040 4510 4940 5340 5710 6050 6380 6690 6980 7270 7560 7820 8080 8320 8560 8800 9030 9250 9470 9680 9900 10100 10300 10490 10890 10880 11070 11940 12770 14290 15620
1% I
179 219 253 283 309 334 357 379 399 418 438 455 473 489 505 521 536 551 565 579 593 606 619 632 645 656 668 680 692 747 799 894 977 6 3405 4165 4810 5380 5895 6370 6810 7210 7600 7970 8330 8670 9000 9320 9630 9920 10210 10500 10770 11020 11300 11550 11800 12030 12290 12510 12730 12980 13200 14250 15220 17020 18610
N o t e : - T h e actual quantities will vary from these'f,'gures, th'e amoun; of variation depending upon the shape of nozzle and size of p,pe ai the point where the pressure is determined. With smooth taper nozzles the actual discharge as about 94% of the figures given in the tables.
F l g . 7A-41
m o s t c o m m o n t y p e s are the r e c t a n g u l a r c o n t r a c t e d w e i r and the 90 ~ V - n o t c h weir. T h e s e are s h o w n in Figure 7 A - 4 2 w i t h the a p p l i c a b l e f l o w f o r m u l a s ,
i m p a c t of the f l o w i n g s t r e a m r e a d s static h e a d + velocity head, and the o t h e r reads the static h e a d only (Figure 7A-43). T h e d i f f e r e n c e b e t w e e n the t w o r e a d i n g s is the velocity head. T h e v e l o c i t y a n d the f l o w are then d e t e r m i n e d from the f o l l o w i n g
g.) P i t o t t u b e
well k n o w n f o r m u l a s .
A pitot tube m e a s u r e s fluid velocity. A small tube p l a c e d in the f l o w s t r e a m gives t w o p r e s s u r e readings; o n e receiving the full
V = C
~
w h e r e C is a c o e f f i c i e n t for the m e t e r determ i n e d by calibration, and hv = velocity head,
274 Process Plant Machinery
.....-~,.?,~.:..--.~; -,~-~~--~_ ~
~
~
t
\ ' i ~" ~ ,""~ "'". '"',-,,~~ " "
Oil-Rectangular Weir With Complete End Contractions
( 1-90" V-Notoh Weir
Q(G.P.M.) = 1495 H~/l (B-O.2H)
Q(G.P.M.) = 1140 He/2
B = Hread H Cest W iniFt"tAFb~ d h i n e e t Weir
H = Head in Feet Above Weir
Fig. 7A-42
Total head -1
i~ ~ J~
Static head
Small holes on both sides of outer tube 7 _~J,~ .....
q,,.,i/(((///.,
.......
~.
Weirs. 6. TPL (TOTAL PUMP LENGTH) - The distance from grade to lowest point of pump. Lift below discharge plus head above discharge plus friction losses in discharge line. This is the head for which the customer is responsible and does not include any losses within the pump.
7. RATED PUMP H E A D -
..............
8. C O L U M N A N D DISCHARGE HEAD FRICTION LOSS Head loss in the pump due to friction in the column Fig. 7A-43
Pltot tube
Capacity = Area x Average Velocity
Since the velocity varies across the pipe, it is necessary to obtain a velocity profile to determine the average velocity. This involves some error, but when properly applied a calibrated pitot tube is within plus or minus 2% accuracy.
Vertical Turbine Pumps Turbine Nomenclature
assembly and discharge head. Friction loss is measured in feet and is dependent upon column size, shaft size, setting, and discharge head size. Values given in appropriate charts in Data Section.
9. BOWL HEAD - Total head which the pump bowl assembly will deliver at the rated capacity. This is curve performance. 10. BOWL EFFICIENCY- The efficiency of the bowl unit only. This value is read directly from the performance
curve.
The horsepower required by the bowls only to deliver a specified capacity against bowl head.
11. BOWL H O R S E P O W E R -
BOWL HP = 1. DATUM OR G R A D E - The elevation of the surface from which the pump is supported. 2. STATIC LIQUID L E V E L - T h e vertical distance from grade to the liquid level when no liquid is being drawn from the well or source. The distance between the static liquid level and the liquid level when pumping at required capacity.
3. D R A W D O W N -
4. PUMPING LIQUID L E V E L - The vertical distance from
grade to liquid level when pumping at rated capacity. Pumping liquid level equals static water level plus drawdown.
5. SETTING- The distance from grade to the top of the pump bowl assembly.
Bowl Head • Capacity 3960 • Bowl Efficiency
12. TOTAL PUMP H E A D - R a t e d pump head plus column and discharge head loss. Note: This is new or final bowl head. 13. SHAFT FRICTION L O S S - The horsepower required to
turn the lineshaft in the bearings. These values are given in appropriate table in Data Section.
Sum of bowl horsepower plus shaft loss (and the driver thrust bearing loss under certain conditions).
14. PUMP BRAKE H O R S E P O W E R -
15. TOTAL PUMP EFFICIENCY (WATER TO WATER) - The
efficiency of the complete pump less the driver, with all pump losses taken into account. Efficiency =
Specified Pump Head x Capacity 3960 x Brake Horsepower
Appendix 7A: Centrifugal Pump Fundamentals 275
.
1
GRACe
.....J
(At
J
I
qlm
gliselm~ IPrmmm Although hydraulic balancing reduces impeller thrust, it also decreases efficiency by 1 to 5 points by providing an additional path for liquid recirculation. NOTE:
16. OVERALL EFFICIENCY (WIRE TO WATER)-The efficiency of the pump and motor complete. Overall efficiency = total pump efficiency x motor efficiency. 17. SUBMERGENCE- Distance from liquid level to suction bell.
Vertical Turbine Pumps Calculating Axial Thrust Under normal circumstances Vertical Turbine Pumps have a thrust load acting parallel to the pump shaft. This load is due to unbalanced pressure, dead weight and liquid direction change. Optimum selection of the motor bearing and correct determination of required bowl lateral for deep setting pumps require accurate knowledge of both the magnitude and direction (usually down) of the resultant of these forces. In addition, but with a less significant role, thrust influences shaft H.P. rating and shaft critical speeds. IMPELLER THRUST
Impeller Thrust in the downward direction is due to the unbalanced discharge pressure across the eye area of the impeller. See diagram A. Counteracting thi.~ load is an upward force primarily due to the change in direction of the liquid passing through the impeller. The resultant of these two forces constitutes impeller thrust. Calculating this thrust using a thrust constant (K) will often produce ony an approximate thrust value because a single constant cannot express the upthrust component which varies with capacity. To accurately determine impeller thrust, thrust-capacity curves based on actual tests are required. Such curves now exist for the "A" Line. To determine thrust, the thrust factor "K" is read from the thrust-capacity curve at the required capacity and given RPM. "K" is then multiplied by the Total Pump Head (Final Lab Head) times Specific Gravity of the pumped liquid. If impeller thrust is excessively high, the impeller can usually be hydraulically balanced. This reduces the value of "K". Balancing is achieved by reducing the discharge pressure above the impeller eye by use of balancing holes and rings. See diagram B.
Although hydraulic balancing reduces impeller thrust, it also decreases efficiency by one to five points by providing an additional path for liquid recirculation. Of even greater concern is that should the hydraulic balancing holes become clogged, (unclean fluids, fluids with solid content, intermittent services, etc.), the impeller thrust will increase and possibly cause the driver to fail. Hydraulically balanced impellers cannot be used in applications requiring rubber bowl bearings because the flutes on the inside diameter of the bearings provide an additional path to the top side of the impeller, thus creating an additional down thrust. Hydraulically balanced impellers should be used as a "last resort" for those situations where the pump thrust exceeds the motor thrust bearing capabilities. DEAD WEIGHT
In addition to the impeller force, dead weight (shaft plus impeller weight less the weight of the liquid displaced) acts downward. On pumps with settings less than 50 feet, dead weight may be neglected on all but the most critical applications as it represents only a small part of the total force. On deeper setting pumps, dead weight becomes significant and must be taken into account. NOTE:
We normally only take shaft weight into consideration as dead weight, the reason being that impeller weight less its liquid displacement weight is usually a small part of the total. SHAFT SLEEVES
Finally, there can be an upward force across a head shaft sleeve or mechanical seal sleeve. In the case of can pumps with suction pressure there can be an additional upward force across the impeller shaft area. Again for most applications, these forces are small and can be neglected; however, when there is a danger of upthrusts or when there is high discharge pressure (above 600 psi) or high suction pressure (above 400 psi) these forces should be considered. MOTOR BEARING SIZING
Generally speaking a motor for a normal thrust application has as standard, a bearing adequate for shutoff thrust. When practical, motor bearings rated for shutoff conditions are preferred. For high thrust applications (when shutoff thrust exceeds the standard motor bearing rating) the motor bearing may be sized for the maximum anticipated operating range of the pump. Should the pump operate to the left of this range for a short period of time, anti-fraction bearings such as angular contact or spherical miler can handle the overload. It should
276
Process Plant Machinery
be remembered, however, that bearing life is approximately inversely proportional to the cube of the load. Should the load double, motor bearing life will be cut to 1/8 of its original value. Although down thrust overloading is possible, the pump must never be allowed to operate in a continuous up thrust condition even for a short interval without a special motor bearing equipped to handle it. Such upthrust will fail the motor bearing. CALCULATING MOTOR BEARING LOAD
As previously stated, for short setting non-hydraulic balanced pumps below 50 feet with discharge pressures below 600 psi and can pumps with suction pressures below 100 psi, only impeller thrust need be considered. Under these conditions:
Where:
Motor Bearing Load (Ibs) 7~mp - " KHL x SG
Impeller Thrust (Ibs) K = Thrust factors (Ibs./ft.) HE = Lab Heat (ft.) SG = Specific Gravity
For more demanding applications, the forces which should be considered are impeller thrust plus dead weight minus any sleeve or shaft area force. In equation form:
(3) Shaft Area Force = Shaft area x Suction pressure * Oil Lube shaft does not displace liquid above the pumping water level and therefore has a greater net weight. CALCULATING AXIAL THRUST
Shaft Dia (in) 1 1 1 1 1 2
3/16 1/2 11/16 15/16 3/16
Shaft Dead Wt. (Ibs/ft.) Open Lineshaft
Closed Lineshaft
2.3 3.3 5.3 6.7 8.8 11.2
2.6 3.8 6.0 7.6 10.0 12.8
Shaft Area (in 2)
Sleeve Area (in)
.78 1.1 1.8 2.2 2.9 3.7
1.0 1.1 1.1 1.5 1.8 2.0
THRUST BEARING LOSS
Thrust bearing loss is the loss of horsepower delivered to the pump at the thrust bearings due to thrust. In equation form: LTB = .0075 ( BHP
where:
Motor Bearing Load = Timp + Wt O) - sleeve force (2) - shaft area force ~3~ = "It (1) Wt. = Shaft Dead Wt. x Setting In Ft. (2) Sleeve Force = Sleeve area x Discharge pressure
LTB = Thrust bearing loss (HP) BHP = Brake horsepower Tt = Motor Bearing Load (Lbs.) = Ttmp + W t (1) - - sleeve force (2) - shaft area force (3)
Vertical Turbine Bearing Material Data Material Description
Temp. and S.G. Limits
Remarks
1. Bronze-SAE 660 (Standard) #1104 ASTM-B-584-932 2. Bronze-SAE 64 (Zincless) # 1107 ASTM-B-584-937 3. Carbon Graphite Impregnated with Babbitt
- 5 0 to 250 ~ Min. S.G. of 0.6 - 5 0 to 180 ~ Min. S.G. of 0.6 - 4 5 0 to 300 ~ All Gravities
General purpose material for non-abrasive, neutral pH service. 7% TinH% Lead/3% Zinc/83% Cu. Similar to std. bronze. Used for salt water services. 10% Tin/ 10% Lead/80% Cu. Corrosion resistant material not suitable for abrasive services. Special materials available for severe acid services and for temp. as high as 650 ~. Good for low specific gravity fluids because the carbon is self-lubricatin$. Corrosion resistant except for highly oxidizing solutions. Not suitable for abrasive services. Glass filled Teflon also available. Used on non-abrasive caustic services and some oil products. Avoid water services as beatings can rust to shaft when idle. Test with bronze Bearings. Excellent corrosion resistance to a pH of 2. Good in mildly abrasive services. 80% Lead/3% Tin/17% Antimony. Use in abrasive water services. Beatings must be wet prior to start-up for TPL 50'. Do not use: For oily services, for stuffing box bushing, or with hydraulically balanced impellers. For services that are corrosive, backing material other than Phenolic must be specified. Expensive alternate for abrasive services. Hardfaced surfaces typically in the range of Rc72. Other coatings are chromium oxide, tungsten carbide, colmonoy, etc. Consult factory for pricing and specific recommendation.
4. Teflon 25% Graphite with 75% Teflon 5. Cast Iron ASTM-A-48 CL30 Flash Chrome Coated 6. Lead Babbitt
7. Rubber w/Phenolic backing (Nitrile Butadiene or Neoprene)
8. Hardened Metals: Sprayed on stainless steel shell (Tungsten Carbide)
- 5 0 to 250 OF All Gravities 32 to 180 ~ Min. S.G. of 0.6 32 to 300 ~
32 to 150 ~
All Temperatures All Gravities
APPENDIX 7B
Change of Performance*
Different industries with many different processes will have requirements for the same pump to operate at different capacities and different heads, and to have a different shape of the head-capacity curves. To ideally satisfy these requirements, one should have a variable-speed pump with adjustable vanes in the impellers. But because most of the drivers in the process industries operate at constant speed, and because the adjustable vanes cannot be produced economically, variable pump performance must be achieved by mechanical means without sacrificing efficiency. In order to provide this flexibility at minimum cost, studies were made to change pump performance within a given pump casing. This can be accomplished by varying the impeller design, cutting impellers, changing the running speed, modifying the impeller vane tips, filing the volute cutwater tip, or orificing the pump discharge. Pump users would prefer to use the same casing for a wide variation of pump performance. The pump casing is usually the most costly part of the pump. To replace a pump casing means extensive and costly work on base plate and piping. The prediction of pump performance by modifying parts other than the casing is based largely on experimentation. Many tests have been conducted by the various pump companies in such areas as: 1. 2. 3. 4. 5. 6.
Trimming the pump impellers Removing metal from the tips of impeller vanes at the impeller periphery Removing metal from the volute tip in the pump casing Providing impeller vanes of the same angularity, but different width Providing impellers with different numbers of vanes and different discharge angles Orificing the pump discharge in the pump casing
We will consider each of these means, but before we do so, we should review the so-called laws of affinity relating to centrifugal pumps. These are theoretical laws or rules that apply to the change in performance of a centrifugal pump by a change in the speed of rotation or a change in the impeller diameter of a particular pump. It should always be remembered in using these laws of affinity that they are theoretical and do not always give exact results as compared with tests. However, they are a good guide for predicting the hydraulic performance characteristic of a pump from a known characteristic. A performance change can be obtained by either the speed of rotation or the outside diameter of the impeller. I. Constant impeller diameter A. The capacity varies directly as the speed GPM1
RPM1
GPM2
RPM2
* Source: Goulds Pumps, Inc., Seneca Falls, NY. Adapted by permission. 277
278
Process Plant Machinery
B. The head varies as the square of the speed Head1 = [RPM1 ] 2 Head2
LRPM2
C. The horsepower varies as the cube of the speed BHPI = [RPM1 ] 3 BHP2
LRPM2
II. Constant speed A. The capacity varies directly with the impeller diameter. B. The head varies as the square of the impeller diameter. C. The horsepower varies as the cube of the impeller diameter. These relationships can be expressed in a simple formula:
Imp,lerioeter, Impeller diameter2
IMPELLER
[ ea l ] r"" l ] GPM2
Head2
LBHP2
CUTS
Assuming that the impeller represents a standard design and that the impeller profile is typically of average layout and not specifically designed for high NPSH, pump performance with trimmed impellers will follow the affinity laws as some vane overlap is maintained. To compensate for casting and mechanical imperfections, correction factors are normally applied to the impeller cuts (Figure 7B-1). The efficiency of the cut impellers (within a 25 percent cut) will usually drop about two points at the maximum cut. On high specific speed pumps, the performance of the cut impellers should be determined by shop tests.
r
A
~
8's
FIGURE
7B-1
~
CALCULATED DIA. IN PERCENT OF ORIGINAL DIA CORRECTION FOR IMPELLER DIA CUT
9:5
,oo
Appendix 7B: Change of Performance 279
O1 " Q x ~
Q Q1 B A
1
= = = =
Capacity Normal Capacity After Underfiling Vane Spacing Normal Vane Spacing After "Underfiling"
Metal removed
Origin
"
/
/
/ /
Head-Capacity After "Underfiling"
Vane thickness
"underfiling"
_j - 021
Cornpres,.-~r dr:vez start-up interlock b'tart of auxihary oil pump
Line symbols
---- bube os letup:, --C o o l . [ : ~ w~te: .... C~ndc[~tc Ir._~runlcn: a~r Elc~tric/,~lcc~rc,mcaiqnal Mochan ic~,l co:,,1~:,-.r Pav,~r c,i] supply Power eli ren:rn
--
_I ...
"v~nt:ine
Seahngair C,.I mist
Abbreviations D V
Drain ','bnI
Appendix l lD: Isotherm Turbocompressors 463
Applications
Flg. 11D-51 Alr compressor, type RIK 56, at the East Drlefonteln mlne of the Gold Flelds of South Afrlca Ltd., Johannesburg, compressing 51 000 m3/h from 0.825 to 9.5 bar. Power Input 4000 kW. (Source: Sulzer Turbo Ltd, Z0rlch, Swltzerland)
Flg. 11D-52 Alr compressor, type RIK 71, Installed In one of AGA's alr separatlon plants In the USA. 65 000 Nm3/h are compressed from 0.97 to 6.76 bar; power Input 5535 kW. (Source: Sulzer Turbo Lid, Z0rlch, Swltzerland)
464 Process Plant Machinery
Fig. 11D-53 Nitrogen compressor train consisting of an Isotherm compressor, type RIK 56, with RZ 3 5 - 6 booster, Installed at Union Carbide's air separation plant Prentiss, Canada. The 7900-kW electric motor driven set compresses 37 420 Nm2/h of nitrogen from 0.91 to 47.1 bar. (Source: Sulzer Turbo Ltd, ZQrich, Switzerland)
Fig. 11D-54 Compact Isotherm process air compressor, type RIK 63, driven by a tall gas expander E 40 and a steam turbine with a combined power output of 8200 kW. The set compresses 50 000 Nm3/h of air to 10 bar in a South Korean nitric acid plant. (Source: Sulzer Turbo Ltd, Z(irich, Switzerland)
Appendix l lD: Isotherm Turbocompressors 465
Flg. 11 D-55 Oxygen plant Burns Harbor, USA, of Union Carblde's Llnde Dlvlslon, each alr separatlon llne supplylng 2000 st/d of oxygen to nelghbourlng steel works. The two motor-drlven alr compressors, type ARI 80, dellver 300000 Nma/h each at 7.2 bar, absorbing 25.2 MW. The two oxygen compressors supply 55 000 Nm3/h each at 34.2 bar; power Input 9050 kW. The maln Items of thls compressor plant are: 1 - Oxygen compressor I; 2 - A l r compressor I; 3 - A l r filter house I; 4 - Alr aftercooler for alr compressor I; 5 - Oxygen compressor II; 6 - Control room II; 7 - Alr aftercooler for alr compressor II; 8 - Electrlc motor II; 9 - Nolse attenuatlng enclosure for gearbox; 10 - A l r compressor II; 11 - O11 reservolr II; 12 - Alr filter house II. (Source: Sulzer Turbo Ltd, Z0rlch, Swltzerland)
Fig. 11D-56 These two air compressors, type ARI 56, Installed ad Anglo American's Vaal Reefs mine, South Africa, are the biggest mining compressors of the world. Each machine supplies 170 000 m3/h at 9 bar and is driven by a 15-MW synchronous motor. (Source: Sulzer Turbo Ltd, Z0rich, Switzerland)
466 Process Plant Machinery
Fig. 11D-57 Skid-mounted air compressor unit, type RIK 56, with R 35-5 and steam turbine arriving at an ammonia plant in Cartagena (Spain). Compressor capacity 34 900 Nm3/h, discharge pressure 34 bar. (Source: Sulzer Turbo Ltd, Z(lrich, Switzerland)
Fig. 11D-58 One of two Isotherm compressors, type ARI 63, Installed In the steelworks of Diefflen/Dillingen, FRG, shown during erection on site. The compact units with Integrated coolers allow their Installation also at locations where space is very limited, still ensuring easy access to all vital parts for maintenance. Each compressor delivers 180 000 Nm3/h of atmospheric air at 7.1 bar. Power Input 15.4 MW. (Source: Sulzer Turbo Ltd, ZQrich, Switzerland)
Fig. 11D-59 This motor-driven. 12.6-MW compressor, type ARI 63, delivers 139420 Nm3/h of air at 7.17bar to Airco-Cryoplants' air separation installation supplying oxygen for the Coolwater coal gasification plant at Daggett, Cal., USA. The eyngaa produced serves as fuel for a 100-MW cogeneration power plant. (Source: Sulzer Turbo Ltd, Z0rich, Switzerland)
APPEN DIX 11 E
Gas Seal Design
Gas seal designs are a critical feature in compressor design, operation and maintainability. Manufacturers constantly seek improvements in this area as is indicated by the following extracts from a technical paper. The purpose of this appendix is to emphasize two of the main technical features incorporated in the design of the D-R Gas Seal: 9 The groove pattern (generates high film stiffness and optimum pressure distribution) 9 The "L" Sleeve design (reduces the risk of "hang up" and therefore cuts operational costs).
Gas Seal Principle Figure 11E-1 shows a simplified cross-section drawing of a D-R gas seal where the main active parts are: 9 The rotating seat, also called tungsten carbide ring 9 The stationary seat, also called the carbon ring 9 The pusher sleeve, also called "L" sleeve Figure 1 I E-2 shows, in parallel, the essential parts of a D-R gas seal, and the detailed representation of the closing and opening forces which determine the equilibrium of the seal faces. When calculating the balance of forces on the carbon face for a given gap, we take into account: 9 the opening forces (OF): sum of (pressures inside the gap x surface) 9 and compare that to the closing forces (CF): sum of (pressures at the back of
the carbon face x surface on which they apply + spring force (Fs) -friction force (FI") If the gap increases, then the opening forces will decrease; if the gap decreases, then the opening forces will increase. The film of gas acts just like compression springs: If the speed increases, then the opening forces will increase as the grooves will generate more lift. Conversely, we can also adjust the closing forces. If we increase the hydraulic diameter, then the closing force will diminish; and if we lower the hydraulic diameter, the closing forces will increase. This is called 'playing with the balance ratio'. * Source: Dresser-Rand, Olean, NY. 467
468
ProcessPlant Machinery
FIGURE 11 E-1
(Source:Dresser-Rand, Olean, NY)
FIGURE 11E-2 (Source: Dresser-Rand, Olean, NY) This can be put in simplified equations as follows CF = Pin • S l + Pout X 52 "+" Fs OF = k (rpm, press, groove geometry, temperature, gap, gas characteristics) In the steady-state situation O F - - C F . Thus the solution of these equations is to calculate the gap (operating gap) which exactly balances the opening forces and the closing forces
Groove Pattern and Pressure Profile Considering the above equations, while it is simple to calculate the closing forces, the calculation of the opening forces is a complex function of several parameters, which are not independent. Therefore Dresser-Rand have developed a computer code that will calculate pressure, speeds and temperature distributions at every point within the interface. This program iterates until the operating parameters are established and stable; in fact it finds the gap at which the opening forces exactly balance the closing forces. Of course various groove geometries can be analyzed and then optimized in order to provide the highest reliability and performance to the gas seal.
Appendix 11E: Gas Seal Design Ideally, a gas seal would like to have a minimum gap in order to minimize the leakage rate (the gas seal leakage is approx, a function of the gap raised to the power 3). However, a smaller gap gives a higher risk of accidental contact between the two faces of the seal. A small gap as well as the highest possible gas film stiffness will be the optimum. A uniform pressure distribution between the faces is important, since it reduces the local deformation of the parts. Therefore, let's discuss pressure profiles, balance of forces, and gas film stiffness. It is clear that the layer of gas trapped between the rotating seat and the carbon face of the gas seal changes pressure as it goes from the OD (at seal supply pressure to the ID (at flare pressure or atmospheric pressure). The grooves in the rotating seat alters the normal pressure decay and generates zones of overpressure. Using the pressure at each point of the interface, calculated with the above mentioned calculation code, it is possible to generate a 3D representation of the pressure profile. One example is shown in Figure 11E-3. From the same computation output it is also possible to show the isobar curves (Figure 11E-4) which highlight the overpressurized areas responsible for the separation of the seal faces. Comparing the various pressure profiles and isobar curves obtained with different groove patterns, it is possible to select a groove shape which provides high pressure rise and as uniform as possible pressure distribution. The patented groove pattern used by Dresser-Rand satisfies these two important criteria. As an example, Figures 11E-5 and 6 show isobar curves at different operating conditions (high speed/low speed and high pressure/low pressure)
Gas Film Stiffness Gas film stiffness is a major parameter for gas seals as it can be used to evaluate the ability of a gas seal to resist sudden positional changes (surge for instance), or also to compare two seals with different groove geometry Let's define what the gas film stiffness is: At a given operating gap an opening force (OF1) is generated within the interface, if the gap is forced to close (or open) by say 1/100 of its value (gap/100), a new opening force will develop (OF2).
FIGURE 11E-3 (Source: Dresser-Rand, Olean, NY)
469
470
ProcessPlant Machinery !:'::.':i i
52.3 50.5 48.8
47 45.2
-0.05
::i. i).. i:':i
41.7 39.9 38.2 36.4
34.6 i:'i :::i~ :i.
0.05
31.1 29.3 27.5 25.8 24 . . . . . . .
0.05
0
-0.05
9. .
20.5
18.7
16.9
15.2 13.4
FIGURE 11E-4
(Source: Dresser-Rand, Olean, NY)
In a similar way, springs are calculated with the formula: dF = S x dX, hence S = dF/dX in this case S = stiffness; dF = ( O F 2 - OFI); and dX = gap/100. The gas film stiffness is ( O F 2 - OF1) x 100/gap. As an order of magnitude, stiffness of more than 3 kN/micron for a medium size of seal (4.875 inch diam) at 50 Bar and 11500 rpm are typical. See Figure 11E-7. In imperial units, this would be about 175 x 106 lbs/in (at 725 psi). This is a very high value, but necessary to avoid any contact between the two faces which at the same time are separated by about 4 microns (0.16 thousandth of an inch). At higher pressure the stiffness values are even higher, which is necessary as the gap will also be smaller. In fact, the stiffness value gives an indication of how the seal can withstand axial forces (especially abnormal forces due to vibrations or upset conditions.) Incidentally, it is possible to use the same program to calculate the behavior of seals having a symmetrical groove pattern (bi-directional). The results demonstrate a significant decrease of gas film stiffness together with a smaller gap. This is why bi-directional seals are not as forgiving as uni-directional seals. The Hang-up Syndrome
Considering again the seal equilibrium, let us analyze now the event of a very low sealing pressure (for instance start up conditions before pressurizing the compressor). See Figure llE-8. When the seal is depressurized, only the spring force can close the seal gap. In this situation, if the shaft of the compressor has a small axial displacement (for
Appendix 11E: Gas Seal Design
0
'PRESdr.RT'I" 54.1 52.3 50.5 48.8 0.04 47 45.2 43.5 0.05 41.7 39.9 0.06 38.2 38.4 34.6 0.07 32.9 31.1 29.3 0.08 27.5 25.8 0.09 24 22.2 20.5 18.7
-0.05
16.9
15.2 13.4
11500 rpm, 50 Bar inlet, ISOBAR 'PRESBV.R~ 52.3 50.5 48.7 48.8 0.04 45 43.2 41.3 0.05 ~.5 37.7 0.06 ~.8 34 0.07 0.08 0.09 1
0
I
-0.05
~.2 ~.3 29.5 ~.7 ~.8
21.2 19.3 17.5 15.7 13.8 12 10.2
500 rpm, 50 Bar inlet, ISOBAR FIGURE 11E-5 (Source: Dresser-Rand, Olean, NY)
instance differential thermal expansion between shaft and compressor casing) the friction between the O-ring and the seal housing or the deformation of the O-ring itself, may prevent the spring closing the seal. A much larger gap (50 to 100 times the normal gap) may then appear between the two faces of the seal. The leakage through this interface could then be so high that pressurizing the unit becomes impossible.
471
472
ProcessPlant Machinery
'PRES150.RT'r' 157 151 146 141 0.04 136 131 125 0.05 120 115 0.06 110 105 99.5 0.07 94.3 0.08 0.09 0
-0.05
89.1 84 78.8 73.6 68.4 63.2 58 52.9 47.7 42.5 37.3
11500 rpm, 150 Bar inlet, ISOBAR 'PRESOO2.RTT' 3.47 3.06 2.65 2.24 0.04 1.63 1.42 0.05 0.06 0.07
0.08 0.09 0
-0.05
11500 rpm, 2 Bar inlet, ISOBAR FIGURE 11E-6 (Source: Dresser-Rand, Olean, NY)
This is known as the 'hang-up syndrome'. The only remedy is to disassemble the compressor end, remove the seal cartridge, fix the gas seal and reinstall it. This, of course involves unnecessary down time and high maintenance costs, usually at a time when there is an urgent need to have the compressor up and running. Dresser-Rand has developed and incorporated in the D-R Gas Seal, an "L" sleeve design (patent pending) which drastically reduces the risk of "hang up".
Appendix 11E: Gas Seal Design
FIGURE 11E-7 (Source: Dresser-Rand, Olean, NY)
FIGURE 11E-8 (Source: Dresser-Rand, Olean, NY)
The 'L' Sleeve Design As explained above, the hang-up situation is caused both, by the tendency of the O-ring to stick or to extrude, and by the simultaneous lack of gas pressure, pressure which in normal operating conditions, is the major contribution to the closing force. The Dresser-Rand "L" sleeve design addresses these two causes:
(a) Reduced risk for extrusion The balance diameter O-ring is located remote from the hot area of the seal, thus it is subject to a lower temperature. The clearance may be better adjusted (smaller)
473
474
ProcessPlant Machinery
since the housing and the sleeve are of the same material (same thermal expansion coefficient and therefore less risk or tendency for extrusion).
(b) Increased closing force due to the gas pressure (small but existing) The installation of the O-ring in the sleeve (as opposed to its installation in the housing) tends to decrease the balance diameter, thus slightly improves the closing forces due to the gas pressure (see Figure 11E-9).
FIGURE 11E-9 (Source: Dresser-Rand, Olean, NY)
Chapter 12 Axial Flow Compressors* As stated in the introduction to the preceding chapter, dynamic compressors are machines in which air or gas is compressed by the mechanical action of rotating components imparting velocity and pressure to the air or process gas. In an axial compressor, as the name implies, flow is in the axial direction, i.e., parallel to the axis of rotation. Axial compressors are basically high-flow, low-pressure machines, in contrast to the lower flow, high-pressure centrifugal compressors. Figure 12-1 shows the performance characteristics of a centrifugal and an axial compressor at constant speed for the same operating conditions. From this figure, a direct comparison of the characteristics is easy. The "turndown" capability of the centrifugal is much larger than that of the fixed geometry axial. The range of operation is greatly increased through the use of variable geometry.
FIELD OF APPLICATION Axial flow compressors have found wide use in refineries, petrochemical plants, and steel mills. Particularly in refineries, applications formerly handled by centrifugal units are now handled by axial flow compressors. This is due to several trends: first, plant sizes are growing dramatically, which brings the air requirements up into a desirable range for axial compressors; second, due to rising energy costs, there exists an increasing trend toward higher efficiencies; and third, technological improvements have made axial compressors more reliable than ever before. Axial compressors are generally more efficient than centrifugal compressors in the common flow range, depending on conditions. An axial compressor will also generally be smaller than a centrifugal compressor designed for the same flow rate. Although the axial flow compressor requires more stages due to the lower pressure rise per stage, the diametral size is much greater in a centrifugal compressor in order to pass the required air flow. The axial compressor must operate at significantly higher speeds for the same condition and is usually more costly than a comparable centrifugal compressor. In applications where speed is not a major consideration, an efficiency and size versus cost evaluation must be made.
Petroleum Refineries These are probably among the largest current users of axial flow compressors for providing the air for catalytic cracking. This service requires 50,000 cfm to 300,000 cfm at discharge pressures from 25 to 50 pounds per square inch gauge (psig). Figure 12-2 depicts a typical installation. * Source: Dresser-Rand Company, Phillipsburg, N.J. 475
476
Process Plant Machinery
FIGURE 12-1 Comparison of axial centrifugal characteristic curves. (Source: Dresser-Rand Co., Phillipsburg, N.J.)
FIGURE 12-2 The first power recovery train in fluid catalytic cracking service in a refinery and the equipment train. The equipment train comprises tandem motor-axial compressor-steam turbine-hot-gas expander units. The compressor is rated at 127,900 inlet cfm with a discharge pressure of 34 psig. The train power rating is approximately 15,000 HP. (Source: Dresser-Rand Co., Phillipsburg, N.J.)
Axial Flow Compressors
Butadiene Plants Air capacities from 60,000 to 150,000 inlet cfm and discharge pressures from 20 to 30 psig with atmospheric intake enable the axial compressors to perform well in this application.
Nitric Acid Plants In plants with capacities in excess of 500 tons per day, axial-centrifugal compressor combinations are frequently used to handle flows from 20,000 to greater than 80,000 cfm (approximately 100 cfm per ton-day) and discharge pressures from 110 to 130 psig for the combined compressor string. Power recovery expanders utilizing process "tail gas" usually drive the compressor string.
Air Separation Plants Axial compressors are used almost exclusively for higher flow air services. Services range to 100,000 cfm and discharge pressure up to 100 psig.
Blast Furnaces Axial compressors are replacing many of the older, less efficient centrifugal blowers. The compressors provide air at discharge pressures from 30 to 90 psig and flows from 125,000 to 350,000 cfm. In addition to the common processes described above, axial compressors are also often used for wind tunnel service, waste treatment facilities, specialized testing facilities, and are being developed for co-generation combustion service.
BASIC AXIAL COMPRESSOR PERFORMANCE CAPABILITIES As described earlier, the axial flow compressor is a machine with wide ranges of capacity, i.e., flow, pressure, and horsepower required. Axial compressor flow capabilities on the low end of the range, say 20,000 to 75,000 cfm, obviously overlap the higher range of centrifugal compressor coverage. It is within this low flow region where cost and size evaluation, along with driver considerations, must control the selection. Above this range, however, axial compressors are often the obvious choice. The physical size of the axial compressor is far smaller than the comparable centrifugal machine that would be required. In many high-flow situations, the axial is a better match for the drivers that will probably be selected. To increase the pressure capability of the axial flow compressors, multiple casing designs have also been developed. These are known as biaxials and triaxials. As their names imply, these are two- or three-body axial compressor trains capable of pressure ratios up to approximately 12 to 1. The machines were developed for use in nitrogen injection services. Horsepower requirements for axial flow compressors range from 3000 HP to 65,000 HP for single casing units. Horsepower inputs vary with the flow and pressure requirements of the service. A simple formula for the approximate power
477
478
Process Plant Machinery requirement of an axial flow compressor would be: Horsepower = where W W R T PR
= = -=
W W x R x T x (PR ~
- 1)
182.5
Wet weight flow, lb/min Gas constant, f t - l b f / l b m - ~ Inlet temperature, ~ Pressure ratio
FUNDAMENTALS OF AXIAL COMPRESSOR DESIGN A multistage axial flow compressor has two or more rows of rotating components operating in series on a single rotor in a single casing. The casing includes the stationary vanes (the stators) for directing the air or gas to each succeeding row of rotating vanes. These stationary vanes, or stators, can be fixed- or variable-angle, or a combination of both. A typical axial flow compressor cross section is shown in Figure 12-3; Figure 12-4 shows an axial compressor with the top half removed. The major components and their nomenclature are depicted in Figure 12-3 for reference use throughout this chapter. There are two basic types of blading that are employed in an axial flow compressor; these are obviously rotating and stationary. A brief overview of these parts is presented next.
Stationary Blades Inlet Guide Vanes T h e first r o w o f s t a t i o n a r y blades is u n i q u e . T h e s e blades are r e f e r r e d to as i n l e t g u i d e vanes. T h e s e vanes are d e s i g n e d to p r o v i d e p r e r o t a t i o n to the air o r gas stream p r i o r to e n t r y i n t o the r o t o r blades. B l a d e p r o f i l e s have a i r f o i l - s h a p e d cross sections.
!
H' II
/
II
-9 Ik
IN I"AKE CASING
r" II/ II
~
(~-"~
]l ////1~
l
/
SURGF
DISCHARGE CASING
DErEcrloN DRIVE
THERMOCOUPL[
RING / -INI.ET GUIDE | /
,.,J i/VANE
) L /
"t_..__
".... .......i]
!1
VARIABLE S T A T O R
ASSEMBLY
1
II
II
II
CASING / FIXEDSTAIO R
!1II
II
[-~,..
~
I HR BEAR
II
GE
. . . . . , , , -,_ _/___ ___ L _.~___l, . . . . . . . .
FIGURE 12-3
/- . . . . . . . / .
.
.
.
.
.
.
.
.
ING
Typical axial flow compressor cross section. (Source: Dresser-Rand Co., Phillipsburg, N.J.)
Axial Flow Compressors
479
FIGURE 12-4 Axial compressor, type AV 100-16, during erection Two identical steam-turbine-driven machines are supplying air to the blast furnace of a British steel works. Suction volume 560 000 Nm3/h discharge pressure 6.2 bar power input 52 000 kW each. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland.) Stator Vanes
The majority of the stationary blades within the compressor are simply called stators. There exist two types of stator vanes, variable and fixed. Variable Stator Vanes. Variable stator vanes fit through the stator casing or a blade carrier of some kind (depending on the manufacturer's design) and are attached to a drive mechanism that moves the vanes with respect to the air flow. A more detailed description of the actuating system is provided below. The inner end of each vane can be shrouded to improve the stress condition and to reduce the interstage losses through sealing strips mounted in the inner shroud. The actuation system used to move the variable stator section is usually a combination of linkages designed to move the vanes simultaneously. One type of linkage system is shown in Figure 12-5. Each variable stator vane is connected to a driving ring by a small link. These rings, one for each stage of blades, are individually connected to a main driving shaft so that the stages move simultaneously. The drive shaft is connected to a hydraulic (or pneumatic) power piston, which, through push-pull effect, opens and closes the stator vane. Different designs are chosen for this system by the various manufacturers. Desirable features would include the following:
1. Solid or "tight" linkage systems to prevent slow or inefficient actuation of the vane position.
480
Process Plant Machinery
FIGURE 12-5 A typical variable stator vane actuation system and linkage arrangement. (Source: Dresser-Rand Co., Phillipsburg, N.J.) 2. A minimum number of "joints" in the system that can wear with time and become loose or seize due to the presence of dirt. 3. Dual power cylinders (one on each side of the unit) to provide even movement of all vanes. Fixed Stator Assembly. The fixed stators are typically welded assemblies comprising the vanes and inner and outer shrouds. These assemblies are fitted into machined grooves in the stator casing. The fixed stator assembly is also fitted with sealing strips for leakage reduction.
Rotating Blades The rotating blades within the axial compressor are appropriately called rotor blades. These are National Aeronautics and Space Administration (NASA)-developed tapered and contoured airfoil sections. The rotor blades have an attachment on one end to allow for assembly within the rotor. A simplified partial section of an axial flow compressor flow path is shown schematically in Figure 12-6. The basic components would typically include the following: 9 An inlet duct to collect and accelerate the gas toward the inlet guide vanes with minimum pressure losses. 9 A row of inlet guide vanes to impart prewhirl to the gas stream in the direction of rotation for smooth entry to the rotor blades and for the control of the inlet relative Mach number. 9 A multiplicity of stages, each consisting of a row of rotor blades and a row of stator vanes of airfoil shape, to increase the static and/or total pressure of the flow.
Axial Flow Compressors
BASIC FLOW PATH
INLET INLET GUIDEVANE
STAGE ~I ~ ROTOR BLAD
--
N
STAGE ~1 STATOR BLADE
STAGE'~n ROTORBLADE~
N
STAGE ~n STATORBLADE
~ EXIT GUIDEVANE I TO DISCHARGE DIFFUSER ROTATION
=--
FIGURF 12-6 Schematic presentation of an axial flow compressor flow path. (Source: Dresser-Rand Co., Phillipsburg, N.J.) The total energy transfer to the gas stream is accomplished by the rotor blades. The hub stagger (the angle between the blade chord and the axis of rotation) is fixed, thereby fixing the amount of work done by each stage and consequently fixing the number of stages necessary to achieve the required discharge pressure. The standard frame design is adjusted to meet the required air flow by varying the rotor and stator blade heights and the unit operating speed. 9 A row of exit guide vanes, oriented to remove the whirl component from the flow leaving the last stage stator vanes, and to begin deceleration of the flow. 9 A discharge diffuser to further decelerate the flow and to convert the residual velocity energy into static pressure rise.
Natural Frequencies and Resulting Stresses Due to the inner and outer shrouding of the fixed stators and the internal shrouding and casing support of the variable stators, vibration occurring at component natural frequencies and the resulting stresses in these components are not of major concern. If a specific manufacturer does not use shrouds on both inner and outer surfaces,
481
482
Process Plant Machinery
it is important to review the frequency analysis of the stationary vanes in the same manner as discussed below for the rotor blades. Since the rotor blades are mounted in a cantilever beam arrangement, i.e., one end unsupported, natural frequency excitation and the resulting stresses must be thoroughly analyzed by the designer. Clear definition of the natural frequencies of the blading, possible sources of excitation within the unit, and the resulting stress levels should be provided by the manufacturer. This should be presented in the form of Campbell and Goodman Diagrams for the compressor blading. (Refer to Chapter 5 for a discussion of these diagrams.) Evidence of the accuracy of the information, in the form of test data or successful long-term operating experience, should be made available by the manufacturer and should be reviewed by the purchaser. Similar scrutiny is appropriate for thrust beating designs. OPERATIONAL LIMITATIONS The successful prediction of the performance of an axial compressor requires knowledge of the flow behavior likely to be encountered in the machine. In certain areas of the operating characteristic, accurate prediction of the performance is not possible. Violent overall instabilities identified as stall, surge, and choke complicate these predictions.
Stall Stall is a commonly used term with regard to axial flow turbomachinery. It is too often incorrectly used as a cause of problems, mainly due to a misunderstanding of the true flow behavior during a stalled condition. Stall occurs when the flow separates from the surface of the blade. An aerodynamic disturbance is formed downstream of the point of separation. Stall is a generalized term. It is often used with additional descriptors to explain the flow condition that causes the separation. For instance, the separation can occur on either side of the airfoil. High positive incidence stall, caused when the angle between the flow and the inlet to the rotor blade is too large, causes separation of the airflow from the suction, or convex, side of the airfoil. High negative incidence stall causes separation of the airflow from the pressure, or concave, side of the airfoil. Conditions of rotating stall can be established when a group of blades becomes stalled. This is a phenomenon of stall cells being created within a stage. As the first blade stalls, it causes a disturbance to the airflow of the adjacent blade, eventually causing it to become stalled. As each subsequent blade becomes stalled, the last blade in the patch of stalled blades begins to recover. Thus, the effect is a rotating patch of stalled blades. Whether a few blades experience stall or a rotating stall cell is formed, the mechanical damage possibly caused by this unstable aerodynamic condition can be significant. It is difficult to determine the actual loads induced on the blades during such an event. Laboratory testing has shown that loading levels can reach ten times normal levels. Surge As with any dynamic compressor, surge occurs when the slope of the pressure ratio versus capacity curve becomes zero. It is associated with the complete breakdown
Axial Flow Compressors
of flow through the machine, and it takes place when several adjacent stages are subjected to high positive incidence stall. At any given speed, as the inlet flow is reduced, a point of maximum discharge pressure is reached. As flow is further reduced, the pressure developed by the compressor tends to be lower than the pressure in the discharge line and a complete flow reversal of an oscillatory nature results. The reversal of flow tends to lower the pressure in the discharge line and normal compression resumes. If no change to either the system back pressure or the operation of the compressor occurs, the entire cycle is repeated. This cycling action is an unstable condition varying in intensity from an audible rattle to violent shock, depending on the energy level of the machine. Intense surges are capable of causing serious damage to the compressor blading and seals. The uncertainties surrounding this oscillating flow are cause for concern.
Surge Control It is standard procedure at process plants to install reliable antisurge control equipment in the compressor piping to prevent operation in the surge region. A typical surge control system should incorporate or encompass recycle loops, i.e., valved bypass piping to provide sufficient flow through the compressor to keep it away from surge. Experience points to the following requirements: 9 The system is to be electronic rather than pneumatic for the fastest response time, and the surge valve must be interlocked with the trip circuit such that it immediately opens on a train trip. 9 The control system logic requires flow, pressure, and stator vane position input. 9 The surge valve positioner-operator system must be capable of driving the valve fully closed to fully open in one second and fully open to fully closed in ten seconds. 9 The surge valve should open on the loss of any input signal or operator medium. 9 The surge valve should be sized to pass full flow at any point along the surge line with the valve at 60% open and with full consideration given to the downstream system pressure drop. The point on the surge line requiring the largest valve and discharge system is normally at maximum speed with stators full open, but a point at lower flow with a lower discharge pressure may in some instances dictate size. 9 A check valve should be installed close to the compressor discharge just downstream of the surge valve connection in the discharge line. In addition to the antisurge control system discussed above, a thermocouple detection system can also be employed to determine the presence of the surge recycling effect. When an axial flow compressor experiences surge, it essentially undergoes a momentary internal gas flow reversal. This flow reversal will slightly elevate the stator casing inlet gas temperature. The increase results from the intermediate gas, which has been heated by compression, flowing back into the inlet area. This detection system should be implemented in addition to a primary antisurge control system. It consists of thermocouples installed in the airstream just upstream of the inlet guide vanes. Upon reaching a predetermined setpoint, which indicates surge, a signal is sent commanding the surge control valves to go to the full open position. Upon reaching an acceptable level of temperature, i.e., the compressor has recovered from the surge, the surge valves are permitted to return to the normal position and control of the surge valve is returned to the primary surge control system.
483
484
Process Plant Machinery ,~
OPERATIONAL LIMITATIONS ABC DEF EBF DGH GHA
c r-w
-
ROTATING STALL, FR()NT STAGES SHOCK STALL, FRONT STAGES CHOKING STALL, DISCHARGE STAGES ROTATING STALL, DISCHARGE STAGES SURGE
SURGE
~
G
,,
o_
o
E
B
Vq..
--~
J
9
CAPACITY
FIGURE 12-7 Operating envelope for axial compressors. (Source: Dresser-Rand Co., Phillipsburg, N.J.) Choked Flow Choking occurs when the slope of the pressure ratio versus capacity curve approaches infinity. It occurs at the point where a further increase in mass flow through the cascade is not possible. Choked flow is associated with the flow velocity reaching a Mach number of 1.0 at some cross section within the machine. Unlike with surge, there is no accompanying increase in noise level or machine vibration amplitude. Choke is a "quiet" phenomenon, which, when operation continues for extended periods of time, can cause damage to the rotating blades, with eventual failure possible.
Choke Control A choke control system is needed to avoid operation within the detrimental region. A typical system would encompass a choke valve designed for minimum pressure drop in the open position and capable of full response in ten seconds. The valve would open in the event of signal failure and would respond to a control system using differential pressure (flow) and discharge pressure as inputs. Surge and choke conditions are affected by geometry, speed, and ambient conditions. Each of the aerodynamic instabilities is most likely to occur within a particular primary region of the performance characteristic. Figure 12-7 shows these various regions. STANDARD MAINTENANCE CONSIDERATIONS Like all machinery, axial flow compressors require both periodic preventive and corrective maintenance. A daily review of key operating data is quite often the best preventive maintenance strategy for axial flow compressors. These data should include various ambient and process parameters that define the operation of the unit.
Axial Flow Compressors
In addition, machinery vibration and bearing temperature data should be logged. While these daily (or weekly) records may fail to tell the entire operating history, they can nevertheless establish trends of operation, i.e., has the present problem been progressing slowly for several days or weeks, or is it a sudden change? Quite often, these records can provide the information essential to avoiding unscheduled shutdowns simply through the establishment of normal trends. Just as important, an abrupt reading might be cause for immediate investigation. Such a situation may call for an unscheduled shutdown to reduce risking catastrophic failure. To maximize the benefits available from these records, it will be necessary to be consistent. Records must be maintained daily, since without the proper parameters being accurately recorded, the validity of the record could rightly be questioned. A typical listing of the parameters that should be included in the daily log is outlined in Table 12.1. It is recommended that the operators discuss the list with the equipment manufacturer and add any items that both parties feel are critical to the analysis of operational fitness of the machinery. Data logging could be done manually or automatically, using either a process computer or a dedicated machinery condition computer. Typical items to visually inspect on a regular (daily) basis are listed below. 9 Check the unit for oil leaks at flanges, instrumentation outlets, vane actuator connections, etc. 9 Observe the operation of the stator vane linkage during a change of its position. Check for binding of any components. Ensure smooth movement. "Jumping" or sporadic movement usually indicates mechanical binding or foreign matter in linkage joints. 9 Listen to the unit for unusual sounds, such as rubbing (seals or blades) or leaking gaskets. If the sound intensity is significant, investigate the probable causes. 9 Check bearing oil drain sight flow indicators to ensure good oil flow through the beatings. Note any significant change in the running level, e.g., much fuller than normal, less flow than normal. TABLE 12.1
Parameters for Condition Monitoring of Axial
Compressors Inlet temperature Inlet pressure Inlet relative humidity (dew point temperature) Unit flow (at inlet or delivered) Discharge pressure Discharge temperature Stator vane position Unit speed Radial vibration amplitude Axial position Surge valve position Choke valve position Radial bearing pad temperature Thurst bearing pad temperature Lube oil supply temperature Lube oil supply pressure Lube oil drain temperatures Lube oil total flow
485
486
Process Plant Machinery
Compressor Internal Cleaning Modem axial flow compressors will operate for long periods between shutdowns. Well though-out metallurgy is essential in the design and manufacture of rotor blade components. External surface coatings are applied to protect the blades from corrosive and erosive attack. Along with the addition of coatings, compressor blade life may be increased by installing inlet air filtration systems. On-line cleaning of internals is usually considered after severe degradation of compressor performance is noted. With a properly sized and operating inlet filtration system, fouling should be minimized. However, on-line cleaning is treating a symptom rather than curing the cause. The problem is more directly addressed through investigation of the air quality entering the compressor. Proper design of the inlet filtration system has always been important to the manufacturer: proper maintenance of the system must become similarly important to the operator. There are several on-line cleaning methods employed by operators depending on the process and the available cleaning methods. The question of whether or not axial flow compressors should be cleaned during on-line operation is a complex one. Due to possible problems with each of the cleaning methods currently used, it is appropriate to consult the manufacturer. Below is a short description of some of the systems currently in use. The possible problems, from both the process operation and machine reliability points of view, are also highlighted. The most effective cleaning procedure for the compressor is a low-speed water/kerosene wash. It is to be performed at approximately 30% of normal speed. The compressor is essentially soaked in the cleaning solution for approximately thirty minutes. A rinsing cycle removes residual cleaning fluid, and returning to full speed effectively dries the intemal components of the compressor. This is technically the safest and most effective cleaning method. It is, however, the least desirable from the operational view, since it requires the unit to be removed from the process for approximately one hour. This is the approximate time required to complete the procedure. A second process employed today is a water/solvent spray wash system. It is to be used at full speed and, in most cases, is not detrimental to the process. This system requires the addition of a spray nozzle assembly into the inlet casing of the compressor unit. Commercially available cleaning fluids are used. Possible problems include the incomplete atomizing of the fluid prior to entry into the compressor. The blading could be damaged by the impingement of large water particles. In addition, pulsations created on the rotor blades from the spray nozzles add another excitation to be considered in the natural frequency and stress analysis. Most importantly, while this system is somewhat effective on the first stage on the machine, the cleaning efficiency is drastically diminished at each successive stage. The cleaning medium is simply centrifuged to the casing wall and has little effect on either the rotating or stationary blades downstream. The third method of on-line cleaning is one that has been employed in similar equipment for some time. This procedure entails the introduction of crushed walnut shells (or apricot pits) into the airstream. These "cleaning media" are introduced into the piping upstream of the inlet casing. They are fed at a rate of approximately 50 pounds every two minutes. The possible problems with such a system are very similar to those encountered with the spray wash system. The solids are centrifuged to the casing wall so quickly that the effectiveness of the system is severely diminished beyond the first stage. In units employing blade coatings, nut shells or pits can cause accelerated erosion of the blade coatings. In any event, strict control of the injection rate will be required.
Axial Flow Compressors 487 Corrective Maintenance Corrective maintenance may become necessary every three to five years. At that time, inspection and replacement of wearing parts is often appropriate. Prior to the inspection shutdown, an inventory of the available spare parts should be performed. Discussions with the manufacturer should take place to determine that proper quantities of spares are available to ensure a complete and timely turnaround of the machine. A typical inspection and replacement shutdown may require approximately two weeks. During that time, typical inspections would include the following: 9 Rotor blade cleaning and magnetic particle nondestructive testing to ensure the integrity of parts if a complete spare rotor is not available. 9 Nondestructive testing of rotor discs, particularly in the area of blade attachment, to check for evidence of stress-related damage. The normal method is dry powder magnetic particle inspection. 9 Liquid penetrant nondestructive testing of stationary blading to ensure the integrity of the welded joints within the assemblies. 9 Beating clearance check on the previously installed beatings as well as the new bearings. Visual and dimensional inspection of the beatings for evidence of rubbing, wiping, and unusual wear. 9 Rotor check balance to correct any unbalance introduced by rotor blade replacement. Rotor tip clearance check to ensure that proper running clearances are established for safe operation. 9 Variable stator vane linkage check to replace any worn bushings or locking/locating pins. 9 Instrumentation check to ensure the operational indicators to be recorded are accurate and available. 9 Coupling inspection for tooth wear on gear couplings and diaphragm check on diaphragm couplings. 9 Axial rotor-to-stator clearance check during reinstallation of the rotor assembly to ensure the proper rotor-to-thrust bearing positioning. 9 Seal clearance check on all shaft and bearing assemblies. The shaft should be inspected on removal from the unit for any signs of seal contact. 9 Inlet filtration system check to ensure the filter elements are clean and secure. In addition, the inlet piping should be inspected from the inside to ensure no loose pieces exist or foreign objects are inside, which could enter the unit and cause damage to the compressor.
SELECTING AN AXIAL COMPRESSOR* (For additional reference material, see also appendices on barrel and isotherm compressors in the previous chapter, as well as environmental factors involved in Environmental Engineering and Management: Sustainable Development for the Power Generation, Oil & Gas and Process Industries, Butterworth-Heinemann 1998). Axial compressors are being used increasingly for applications which not long ago were clearly considered the domain of centrifugal machines. Thanks to their high specific flow capacity, the corresponding low weight, reduced space requirement and particularly their high efficiency, the axial compressors play a major * Source:Sulzer Turbo Ltd, Zurich, Switzerland. Adapted with permission.
488
Process Plant Machinery
role in the reliable and economic operation of modem, large-scale industrial plants. They now form a vital and indispensable integral part of installations like blast furnaces, air separation plants, fluid catalytic cracking units, nitric acid plants, jetengine test facilities, thermocompression, liquefied natural gas (LNG) and synfuel processes. The present-day technique of modem axial compressors is based on decades of experience. Thousands of units are in continuous operation in industrial plants all over the world, to which thousands more serving as combustion air compressors of gas turbines have to be added. The power input ratings of the industrial applications vary between 2000 and 90000 kW. With a view to economical manufacturing and stocking of the major components, the range of compressors has been standardized. The stocking of major components facilitates prompt delivery of machine parts such as rotor blades and stator blades, bearings, joints, etc., for service requirements. The systematic design of components over the whole range of sizes enables the compressor to be adapted optimally to the required operating conditions. Measurements conducted on the blading of various designs and sizes ensure exact conformity of the design data with the operating conditions.
Product Range The axial compressors manufactured by one of the major producers, Sulzer, come in two types: Type A Compressors with fixed stator blades (FIXAX) Type AV Compressors with adjustable stator blades on all or only some stages at the inlet (VARAX) Each type consists of 12 geometrically graduated sizes with rotor diameters extending from 40 to 140 cm. This completely covers a suction volume range of 70 000 to 1250 000 m3/h. The required compressor size and number of stages, together with the corresponding standardized overall length, are selected according to the suction volume and the thermodynamic head.
Axial Flow Compressors
Type A (FIXAX) The A-type FIXAX compressor is generally used whenever the driver is a steam turbine, a split-shaft gas turbine or a variable-frequency high-speed synchronous motor. The required operating points can be attained by speed variation, and there is basically no need for adjustable stator blades. Fixed-blade machines are also selected for installations where only minor flow variations are required, or if the mass flow is adapted by variation of the suction pressure as in aerodynamic test facilities, for example.
Type A V (VARAX) The AV-series with adjustable stator blades permits a large stable operating range at constant speed. It is therefore used for constant-speed electric motor drive. Nevertheless, this type is being increasingly preferred for steam turbines and singleshaft gas turbines as well. In this particular case, the stator blade control either facilitates operation with limited speed control range (increased reliability of operation for certain turbine types) or, in combination with the speed control, provides an additional extension of the operating range and an improvement of the overall efficiency at part load. Furthermore, it offers the advantage of quick adaptation of the compressor to changed operating conditions without acceleration of the s e t - a characteristic which is of great interest for the periodic charging of air heaters in blastfurnace blowing plants. Roughly 50% of the major axial compressors sold since 1960 have been equipped with adjustable stator blades.
Stator Blade Setting with Electric Servomotor For a great number of processes, the reference value of pressure or mass flow is selected at the process control panel and transmitted to the compressor servomotor. In this case this servomotor is of the electric type with the additional possibility of manual adjustment of the stator blades.
Stator Blade Control with Hydraulic Servomotor If the process calls for automatic pressure or mass flow control, the stator blade adjusting mechanism will be operated by a synchronized pair of hydraulic servomotors.
489
490
Process Plant Machinery
With the exception of the stator blades and their adjusting mechanism, the same standardized construction elements are used for both FIXAX and.VARAX types. Performance Data The following diagrams (Figures 12-8 to 12-11) facilitate the selection of Compressor size 9 Nominal diameter 9 Number of stages Power input Speed Discharge temperature Using Capacity Suction pressure Suction temperature Relative humidity of the air or gas Discharge pressure Molecular mass Isentropic exponent Compressibility factor The following factors and symbols are also used for the calculation: Suction volume (actual) Absolute humidity Polytropic efficiency Indices Suction branch Discharge branch Dry Wet Information Information concerning the selection and performance calculation of an axial compressor is provided later in this section.
D (cm) z(-)
P (kW) n (rev/min) T2 (K)/t2 (~ m (kg/s)
pl (bar abs) Tn (K)/h (~ ~o, (%) P2 (bar abs) M (kg/kmol) k (-) z
(-)
V1 (m3/s) x (-) 1 2
t f
Axial Flow Compressors
Important
Type
491
Experienced engineering personnel are always needed to deal with customer inquiries and optimization of equipment layout.
designation
Figures 12-10 and 12-11 are valid for air at suction conditions l bar, 20~ 70% relative humidity. (Source: Sulzer Turbo Ltd., ZUrich, Switzerland) Figures 1212 and 12-13 describe operating characteristics for type A and type AV compressors respectively. Figure 12-14 is a longitudinal cross-sectional view of an AV compressor. Depending on the specific process requirements, such as extremely high turndown, maximum efficiency within a certain range, flat or steep p - V curve, etc.,
FIGURE 12-8 Determination of the absolute humidity x and the molecular mass Mf of the wet gas. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
492
Process Plant Machinery
FIGURE 12-9 Determination of the discharge temperature t2. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland) various parameters of the blading may be changed in order to adapt the characteristics to such special conditions.
Design Features 9 Robust design with cast casing and separate blade carrier 9 Casing supported by means of pendulum supports (minimum expansion forces) or feet with keyways for the smaller frame sizes 9 Solid or hollow rotors, very smooth running characteristics with integrated balancing pistons 9 Blading with optimal aerodynamic characteristics- with high efficiency, large specific capacity, low stressing, favourable control characteristics 9 Great safety against blade vibrations thanks to the careful selection of the blade profiles and the special blade fixation 9 Adjustable stator b l a d e s - of standard design - for optimum flow control
Axial Flow Compressors
493
FIGURE 12-10 Type A (FIXAX). Selection of the compressor size: nominal diameter D (cm) = frame size as a function of the actual wet suction volume flow V f, (m3/s). Determination of the approximate number of stages z, nominal speed n (rev/min) and power input P (MW). (Source: Sulzer Turbo Ltd., Zi~'rich, Switzerland)
FIGURE 12-11 Type AV (VARAX). Selection of the compressor size: nominal diameter D (cm) =frame size as a function of the actual wet suction volume flow Vf, (m3/s). Determination of the approximate number of stages z, nominal speed n (rev/min) and power input P (MW). (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
9 Maintenance-free, oil-free stator blade adjusting mechanism which is also protected against the ingress of contaminants 9 Possibility of fitting various types of bearings 9 Possibility of fitting various shaft seals
494
Process Plant Machinery
FIGURE 12-12 Operating characteristics for a compressor, type A, with fixed stator blades and variable speed: NP-Reference point (100%)- Design point; P2/Pl- Pressure ratio. Valid for: constant gas data, constant inlet temperature. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
FIGURE 12-13 Operating characteristics for a compressor, type AV, with adjustable stator blades and constant speed: N P - Reference point (100%)- Design point; P2/Pl - Pressure ratio; a - Angular setting of the stator blades. Valid for: constant gas data, constant inlet temperature. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
Axial Flow Compressors
TABLE 12.2 Selection and performance calculation of an axial compressor
calculation example Air compressor, type A
Given Capacity Suction pressure Suction temperature Relative humidity Discharge pressure Dry molecular mass Isentropic exponent Cp/Cv Compressibility factor
rilt = 146.2 kg/s P l = 1 barabs T1 = 3 0 8 K, tl = 3 5 ~ tpl = 60% P2 = 4.5 bar abs Mt = 28.95 kg/kmol k=l.4 Z=I
Calculation instructions 1. Determination of the absolute humidity x and wet molecular mass Mf (from Tl, Pl, tpl) with Figure 12-8
x =0.021 Mf = 28.6 kg/kmol
2. Calculation of the wet mass flow lilf = riat(l + x)
lilf -- 146.2(1 + 0 . 0 2 1 ) = 149.3 kg/s
3. Determination of the actual suction riaf. T1 [m3/s] volume '~/1 = F * . M f . p l
Vl = 133.7 m3/s
4. Determination of the discharge temperature t2 with Figure 12-9
t2 = 225 ~
5. Selection of the compressor size (nominal diameter D) and number of stages Z, determination of speed n and power input P with Figure 12- l 0
D - - 9 0 cm Z=I2 n -- 3600 rev/min P --- 29 M W
6. Type designation
A 90-12
7. Selection of compressor size AV with Figure 12-1 ! 7.1 Steam turbine drive, i.e. no speed restriction (speed ns = nn" x / Z t h / Z s ) = 3926 .x/10.6/11.0 Type designation
D-90 cm Z--II n = 3854 rev/min P = 29 M W AV 90-11
7.2 Electric motor direct drive 3600 rev/min
D = 100 cm Z=II n = 3600 rev/min P = 29 MW AV 100-11
Type designation 7.3 Electric motor direct drive 3000 rev/min (an) 2 (3533) Zs=Zth" -= 10.6. 9
D = 100 cm Z=I5 n = 3000 rev/min
as
P = 29 MW = 14.7 Type designation nn ns Zth Zs
= = = --
AV 100-15
Nominal speed Selected speed for electric motor direct drive or corrected to arrive at a whole number of stages Theoretical number of stages on the basis of Figures 12-10 and 12-11 Selected number of stages to arrive at a whole number or to account for reduced speed in example 7.3. The reduced rotational speed imposed by the speed of the two-pole motor reduces the tip speed proportionately. This must be compensated by increasing the number of stages to produce the desired pressure ratio. * Factor F, if mass rnt given in Nm3/h = 33505 SCFM (14.7 psia, 60F) = 20844 SCFM (14.7 psia, 70F) = 21245 kg/s = 12.027
495
496
Process Plant Machinery
FIGURE 12-14
Longitudinal and cross-sectional view of an A V compressor. (Source: Sulzer Turbo Ltd., Ziirich,
Switzerland)
Casings, Bearing Pedestals Depending on the type of gas and the design pressure, the casings are made of grey cast iron, nodular cast iron or cast steel. The cast design facilitates a rigid construction, effective noise attenuation and aerodynamically favorable layout of the respective ducting. The suction and delivery branches are usually routed vertically downwards. In cases where, due to the composition of the gas and/or the pressure level, steel casings are mandatory, a welded construction can be supplied. The suction branch may then be axial, or both suction and discharge may be routed upwards or downwards. In the vertical central plane, the casing is aligned by two keyways; it is equipped with four supporting feet. It is fixed at one end in the axial direction by one pair of feet. The other pair of feet of large frame sizes rests on pendulum supports with spherical contact surfaces. As a result of this, the casing can take up the thermal expansion in both the axial and lateral direction without difficulty. This feature is particularly advantageous in the case of light steel foundations. On small frame sizes the pendulum supports are replaced by sliding keyways.
Axial Flow Compressors 497
Blade Carrier, Casing Inserts The blade carder inserted in the casing is centered on both the suction and discharge side, and is able to expand freely in the axial direction. The diffusor and the gland inserts are also fitted as separate parts in the casing. The double-casing design with outer casing and blade carder offers various advantages: 9 Rigid casing construction; the clearances in the blade duct are not influenced directly by external pipe forces. 9 Simple fitting of the blades and assembly of the casing parts; the top half of the casing can be raised without dismantling the blade-adjusting mechanism. 9 Possibility of fitting different blade carriers, for adapting the blade duct and thus the compressor characteristics to greatly changed operating conditions. 9 Optimal protection of the adjusting mechanism in the space between the casing and blade carder; the space is kept under suction pressure in order to safeguard the adjusting mechanism against condensation and corrosion attack.
Rotor The rotors are usually of forged solid design. In the case of larger machines or if the moment of inertia must be minimized to limit the power requirement when running up with an electric motor, welded hollow rotors may be used. Integrated balancing pistons at both ends of the rotor facilitate an equalization of the axial thrust. The careful balancing of the rotor at full speed results in highly smooth running characteristics. If necessary, balancing can also be effected in the casing. The labyrinth strips are caulked in the rotor.
Blading Blading with a high degree of reaction, i.e. the increase in pressure takes place exclusively in the rotating components, is employed for the compression of lighter gases such as helium or hydrogen. (See Figures 12-15.) For all other applications, such as the compression of air, blading with a lower percentage reaction is adopted. The increase in pressure is distributed to the rotor and stator blades. This enables the following major advantages to be realized: 9 Higher efficiency with lower aerodynamic loading 9 Widest possible control range with high part-load efficiency at constant speed 9 Largest possible suction volume at given speed 9 Increased reliability of operation thanks to larger radial blade clearances and the omission of guide vane sealings 9 Steeper pressure volume characteristics, especially suitable for capacity control, for the parallel operation of different compressors, for refrigeration processes and for exact adjustment of the blow-off line. The rotor and stator blades are normally made of 13% chrome steel and machined. When handling highly contaminated aggressive air or corrosive gases, alloys with higher chromium and nickel content may be used. The rotor blades have rhomboidal roots and are firmly braced in an exactly defined position in the peripheral grooves of the rotor. This is of particular importance for their vibration-related
498
Process Plant Machinery
FIGURE 12-15 Adjustable stator blade, rotor blade and fixed stator blade with intermediate piece. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland) design. The fixed stator blades are provided with a rectangular foot. The adjustable stator blades are made in one piece with a cylindrical shaft. The latter is seated in a beating bush in the blade cartier. The high damping characteristics of this seating arrangement practically excludes the occurrence of dangerous vibration amplitudes associated with the stator blades.
Stator Blade Adjusting Mechanism for the AV types The adjusting mechanism is located in an annular space between casing and blade carrier, and is thus well protected against contaminants and moisture. It is maintenance-free and does not require any lubrication. Servomotors.
Two types of servomotors are available (see Figures 12-16 and 12-17):
Automatic mass flow, volume or pressure control. The adjusting mechanism is operated by means of two hydraulic servomotors which are affixed laterally to the casing. One of the servomotors is equipped with a positioning transmitter and the second operates hydraulically in parallel. The linear movement of the servomotor piston rods is transmitted directly to the adjusting cylinder by way of two ball and socket joints. Remote setting o f reference value. One single electric servomotor is attached laterally to the bottom half of the casing. Its driving shaft actuates a pivoted fork positioned on either side of the casing in maintenance-free beatings. This fork in turn moves the adjusting cylinder in axial direction. Adjusting cylinder. The adjusting cylinder of welded design can move in the axial direction and is dry-seated. There is no restriction of heat expansion in any direction. U-shaped guide tings are provided on the inner side in which the adjusting levers are engaged. Stator blades. The adjusting levers provided on the end of each stator blade shaft are connected to the guide tings of the adjusting cylinder by means of pivoting sliders. The axial movement of the cylinder is converted into a rotating movement of the stator blades (see Figure 12-18).
Axial Flow Compressors
FIGURE 12-16 Electric servomotor. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
FIGURE 12-17 Hydraulic servomotor. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
The self-lubricating bearing bushes of the blade shafts are seated in the radial holes of the blade carrier. O-ring packings prevent the ingress of contaminants into the stator blade seating.
Shaft Seals Labyrinth seals (Figure 12-19) are used for the standard models. The stainless steel labyrinth strips are caulked on the rotor and are easily replaceable. In case of rubbing due to unbalance, the friction-induced heat is immediately passed to the massive stator, thus avoiding distortion of the rotor. Gas-tight shaft seals and standstill seals can be fitted for special requirements.
FIGURE 12-18 Stator blade adjusting mechanism. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
FIGURE 12-19 Labyrinth shaft seals. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
499
500
Probess Plant Machinery
TABLE 12.3
Material Table, Valid for Air and Similar Gases
Machine component
Material
DIN No.
ASTM No.
Casing Blade carrier Inlet pieces Diffusors Stator blades Adjusting cylinder Journal bearing Rotor Rotor blades
Grey cast iron Nodular cast iron Grey cast iron Grey cast iron Chrome steel Boiler plate Steel and white metal Alloyed steel Chrome steel
GG-25/1691 GGG-42/1693 GG-25/1691 GG-25/1691 X 20 Cr 13/17440 H 1/17155 LgSn 89 28 Ni Cr MoV 85 X 20 Cr 13/17440
Class 35/A48 SA 395 Gr. 60-40-18 Class 35/A48 Class 35/A48 AISI 420 Class A + B/A 285 B 23 A 469-71, Class 6-7 AISI 420
Journal and Axial Bearings Journal bearings. In the normal version, i.e. with the compressor rotor solidly coupled and the rotor thrust transferred to the axial thrust bearing of the prime mover or the gear, the bearing housings are equipped only with journal beatings. Twolobe bearings are provided for the lower speed range; tilting pad journal bearings (Figure 12-20) being generally used for the higher speeds of the smaller frame sizes for reasons of stability. The slight curvature of the adjusting plates allows the bearings to be set accurately on erection. The bearings (Figure 12-21) are firmly held in position by the bearing housing top half. Two-lobe bearings are suitable for both senses of rotation, while tilting pad bearings are essentially for only one direction, although they can tolerate running backwards with a somewhat reduced load capacity. Axial thrust bearings. If a flexible coupling is selected between driver and driven machine, the bearing housing can accommodate the necessary tilting pad thrust bearing. The purpose of this beating is to absorb the remaining thrust of the machine and any significant axial friction thrust of the coupling due to sharp temporary differential expansion between rotor and casing. To provide easy access and reduce the overhang, it is preferably mounted on the free shaft end. The tilting pads are supported on load equalizing segments which allow angularity of the shaft of up to 0.3%. Because the tilting pads are supported eccentrically, thrust bearings are suitable for only one direction, but tolerate a reversed rotation at a somewhat reduced load capacity. Figure 12-22 shows a popular thrust bearing.
Solid coupling It is Sulzer's normal practice to make extensive use of solid couplings allowing the use of only one axial thrust bearing for single- or multiple-casing arrangements. An intermediate shaft, flexible enough to allow for considerable misalignment, is inserted between the two shaft ends of the machines to be coupled together (Figure 12-23). In case of motor-driven units, the normal technique is to use single helical gears provided with thrust collars on the pinion shaft, as shown in Figure 12-24 and 12-25. The thrust collars not only neutralize the axial thrust created by the meshing of the teeth cut at an angle to the axis of the shaft, but also transmit the unbalanced axial thrust of the high-speed rotor train to the thrust bearing on the low-speed wheel shaft. Good gear meshing requires parallelity of gear and pinion shaft and automatically assures parallelity of the contact surfaces of thrust collar and wheel rim. The slight tapering of the thrust collars is responsible for the formation of a wedge-type oil film creating a pressure zone spread out on an enlarged surface with a pressure distribution very similar to that of a standard-oil-lubricated joumal bearing.
Axial Flow Compressors 501
FIGURE 12-20 Multi-segment journal bearing with four tilting pads. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
FIGURE 12-21 Two-lobejournal bearing. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
The relative motion between the two contact surfaces of the thrust collar system is a combination of rolling and sliding and takes place near the pitch circle diameter, resulting in a very small relative velocity. The thrust transmission is therefore effected with almost no mechanical losses. The considerably reduced losses of the single thrust bearing on the low-speed shaft as compared with the high losses of individual thrust bearings on the high-speed train lead to a substantial power saving. Moreover, this low-speed bearing can be more amply dimensioned to provide a much higher overload capacity. For direct turbine-driven compressor trains, the thrust bearing is usually located in the turbine. Also in this case solid couplings with flexible intermediate shafts are much preferred. This coupling arrangement avoids heavy overhung gear couplings which are usually responsible for not clearly defined lower critical speeds and for the phenomena of torque lock leading to additional loading of the axial thrust beating. The resulting axial friction forces can become quite substantial if insufficient attention is given to the cleanliness of the lubricating oil.
Safety Systems and Process Control Two main requirements are to be met by the control system of turbo-compressors. 9 Safety: to prevent the compressor from operating in an unstable range or at other hazardous conditions. 9 Process: to adapt the compressor performance to the demands of the process Sulzer offers the engineering and the supply of complete compressor control and safety systems. This tradition ensures the optimum protection of the compressor and the plant. The control systems may be pneumatic or electronic with hydraulic or electromechanical servomotor. In cases where the compressor control system is engineered or furnished by others, it is Sulzer's practice to review and approve the system in order to ensure the compatibility of all equipment and functions. Thanks to the strict employment of standard signals, the control system can be integrated into other systems without difficulty: it allows remote control, automation of starting and stopping and integration with process computers.
502
Process Plant Machinery
FIGURE 12-22 Kingsbury-type thrust bearing with self-equalizing pads with directed lubrication. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
FIGURE 12-23 The Sulzer solid quill-shaft coupling conforms to API 671 standard and consists of the quill shaft and the two hubs hydraulically fitted onto the shaft ends of the connected machines. On each coupling side, an equal number of tie bolts for axial fixation and tapered dowel pins for torque transmission and centering assure a clearly defined connection. Balancing as a complete assembled unit and correlative marking enable removal and remounting of this intermediate shaft with the connected rotors remaining in place, without affecting the balancing quality and vibration behaviour of the complete string. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
Axial Flow Compressors
..~..~: ............. F:;;.4~,,'., ........................
F,z
/
;"
Cross-section of thrust collar
A-A
F I G U R E 12-24 Method of axial thrust transfer in a single helical gear with thrust collar. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
FIGURE 12-25
1|
._
II
"
Transfer of external forces. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
Safety systems. Antisurge control Axial compressors have a limited stable operating range, regardless of the type of blading and other influencing factors. This range is given by the characteristic curves limited by the surge lines. Surge conditions, occurring on the left side of this surge line, are avoided by an antisurge control system. It measures flow and pressure and can be designed to closely follow the actually measured surge line at a given safety margin. As soon as the operating point approaches the surge line, the controller starts opening the antisurge valve according to the pre-set values (L) (see Figure 12-26).
503
504
Process Plant Machinery
FIGURE 12-26 Characteristics of a turbocompressor p - Discharge pressure; V - Flow rate; C-Compressor characteristic curves; S - Surge line; L - Limit flow (Source: Sulzer Turbo Ltd., Ziirich, Switzerland) For air compressors, the excess capacity is blown off to the atmosphere. For a gas that cannot be wasted to the atmosphere, antisurge control is a bypass control, the unwanted flow being returned to the suction side. A bypass cooler may then be required. The antisurge control system is not a flow or pressure control device, but a safety device which has to act independently of any other control. Vibration, temperature, pressure, power limitation Under certain circumstances external influences may lead to undesired changes of the normal level of vibration, gas and bearing temperatures, pressure and power. A reliable interlock, alarm and shutdown system must protect the compressor and driver from possible damage under such conditions. Auxiliary component control Auxiliary component control assures a safe supply of lube, control and seal oil. Process control. Suction pressure: Constant suction pressure to adapt the compressor flow to an upstream production unit or to maintain a constant evaporation temperature for refrigerating units.
Discharge pressure: Constant discharge pressure in cases where chemical reactions have to take place at a clearly defined pressure, or where the compressor flow has to be adapted to a fluctuating downstream demand. Flow: Constant mass flow control corresponding to a constant plant output.
Process Control System (see Figure 12-27) System for either capacity control and pressure limitation or pressure control with capacity limitation. The capacity is measured by means of a Venturi tube. The computer FY receives input signals from the differential pressure transmitter PDT, pressure transmitter PT as well as the temperature transmitter TT and calculates the actual capacity value for the flow controller FC.
Axial Flow Compressors
The flow controller FC and the pressure controller PC are each equipped with one manual station HIC for the reference value. By way of the minimal-selection relay UY 2, the volume or pressure controller acts upon the positioning controller GC of the stator blade adjusting mechanism. The positioning circuit for the stator blade adjusting mechanism comprises the positioning controller GC, the electrohydraulic converter GY and the position feedback transmitter GT.
Load Limit Control Malfunctioning in the steam or condensate system of the driving turbine may lead in certain circumstances to an undesired drop in speed. In such cases, the stator blades will be closed to such an extent by the load limit controller SC2 that the speed will remain nearly constant. The speed controller SC 2, which is equipped with the reference value station HC, receives its input signal from the speed transmitter ST and has priority over the process controller with minimal selection relay UY2, when necessary.
Antisurge Control System System to maintain stable operation of the compressor, even when the process operating point moves into the unstable range of the compressor performance characteristic. The pressure transmitter PT is used to determine the actual value. A differential pressure transmitter PDT and a function generator UY 3 are employed for setting the reference value. This computer enables the response line (blow-off line) to be
FIGURE 12-27 Antisurge control system and combined discharge pressure~power limit control system: SC 1 - S p e e d controller; E/HGY-Electrohydraulic converter; P D T - Differential pressure transmitter; T T - Temperature transmitter; P T - Pressure transmitter; G T - Position feedback transmitter; S T - Speed transmitter; F Y - Computer; G C - Positioning controller; SC 2 - Load limit controller; UY 3 - Function generator; F C - Volume controller; P C - Pressure controller; B C - Antisurge controller; UY 2 - Minimum-selection relay; H C - Reference value station; HICManual station. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
505
506
Process Plant Machinery
well adapted to the surge limit. The output signal of the antisurge controller BC acts by way of the minimum selection relay UY 2 on the positioning controller GC of the blow-off valve. The valve can also be opened by the manual station HIC and override the antisurge controller. The positioning circuit for the blow-off valve comprises the positioning controller GC, the electrohydraulic converter GY and the position feedback transmitter.
Monitoring Logic The monitoring logic supervises various controller reference values, the speed of the set, time-dependent operations such as start up and the positioning of the final control elements. The system facilitates the determination and processing of any disturbancies in the final positioning elements and transmitter signals.
General System Layout The control system of axial compressors is designed for all types of drive and the most diverse process applications. It corresponds to the latest knowledge and experience in the field of electronic control. 9 9 9 9
Transmitter according to the 2-wire system; standard signals 4 - 2 0 mA Isolating amplifier in the case of external measurement data Plausibility supervision for the determination of measurement data faults Output tracking for automatic changeover to manual control made in case of measurement failure 9 Impulse technique of the control units 9 Positioning elements driven by hydraulic servomotors and controlled by electrohydraulic converters 9 Two changeover contacts for each output for alarm and control purposes.
Features of the System 9 9 9 9 9 9 9
Proven building block system for compressor installations Central simulation and control unit Integrated monitoring system acting directly on the final positioning elements Monitoring of important values according to the "20-of-3 system" Three-week tests on load at works under elevated temperature and supply voltage Electronic fuse protection for each individual functional group Optimal adaptation of the system to the individual control parameters, thanks to the modular system 9 Self-contained functional units as plug-type modules 9 Easy extension and modification at any time thanks to the clearly defined wiring plans 9 Simple interface in the case of the use of process computers.
Typical Compressor Plant P & I Diagrams Figure 12-28(a) depicts automatic control by means of adjustable stator blades. Figure 12-28(b) depicts remote setting of the adjustable stator blades.
Axial Flow Compressors
FIGURE 12-28(a) Automatic control by means of adjustable stator blades
List of major components for automatic (Figure 12-28(a)) 1 Driving motor 2 Gear coupling, oil-filled 3 Speed-increasing gear with thrust collars, thrust bearing on the slowrunning shaft 4 Main oil pump, laterally flanged to the gear-box, driven by the slow-running gear shaft 5 Solid coupling 6 Turning gear 7 Axial compressor with adjustable stator blades 8 Labyrinth glands 9 Servomotor of the stator blade adjusting mechanism 10 Two-stage air intake filter consisting of: 11 Inertia dust separator (lst stage) 12 Dust evacuation fan 13 Roll-band filter (2nd stage) with driving motor and automatic forward feed control
control of adjustable stator blades
15 16 17 18 19 20 21 22 23 24 25 26 27
14 Bypass flap with open alarm switch GAO Suction silencer Non-return valve with air-operated closing cylinder and solenoid valve Antisurge valve, hydraulically operated Venturi tube for measuring the air flow Discharge silencer Blow-off silencer Noise-attenuating hood Ventilation fan with silencer Lube oil supply unit Control oil supply unit Electronic control system and monitoring logic Discharge isolating valve Jacking oil pump
507
508
Process Plant Machinery
FIGURE 12-28(b)
Remote setting of the adjustable stator blades
List of major components for remote 1 Driving motor 2 Gear coupling, oil-filled 3 Speed-increasing gear with thrust collars, thrust bearing on slowrunning shaft 4 Main oil pump, laterally flanged to the gear-box, driven by the slow-running gear shaft 5 Solid coupling 6 Turning gear 7 Axial compressor with adjustable stator blades 8 Labyrinth glands 9 Electric servomotor and gear for remote setting of stator blade adjusting mechanism by means of controller HIC; gear with local manual intervention facility 10 Two-stage air intake filter consisting of: 11 Inertia dust separator (1 st stage) 12 Dust evacuation fan 13 Roll-band filter (2 nd stage) with driving motor and automatic forward feed control 14 Bypass flap with open alarm switch GAO
setting of adjustable stater blades (Figure 12-28(b)) 15 Suction silencer 16 Non-return valve with air-operated closing cylinder and solenoid valve 17 Antisurge valve, pneumatically operated with air failure to open safety device (pressure accumulator) and solenoid valve. The antisurge valve is controlled via an electric/pneumatic converter E/P by the antisurge controller BIC. Manual control of the valve is possible in the sense "Valve to open" (the antisurge controller thereby being overridden) by operating the manual loader HIC 18 Function generator, computing the set points of the antisurge controller BIC in function of the aspired air flow rate 19 Discharge silencer 20 Blow-off silencer 21 Noise-attenuating hood 22 Ventilation fan with silencer 23 Lube oil supply unit 24 Discharge isolating valve 25 Jacking oil pump
Axial Flow Compressors
FIGURE 1 2 - 2 9
Typical oil supply schematic
Scope of supply and functional description for oil supply components (Figure 12-29) 1 2 3 4 5 6 7
8
9
10
11
12
Lube oil tank with auxiliary equipment Electric oil heaters Oil mist fan Filling sieve and breather Degasifier plates Suction strainers Safety arrangement (4 non-return valves) in the main oil pump suction and discharge lines to prevent backflow of oil if the compressor and hence the main oil pump should accidentally turn in reverse direction Auxiliary oil pump, electric-motordriven; automatic start up of the pump in the event of oil pressure dropping Twin oil cooler, each of the coolers being sized for full flow, with transfer valve for cooler changeover during operation Oil temperature control valve, maintaining constant the oil cooler oil outlet temperature by bypassing more or less the oil around the cooler Oil pressure control valve, maintaining constant the lube oil supply pressure Twin oil filter, each of the filter screens being sized for full flow,
13
14 15 16 17 18 19
20
21
22
with transfer valve for filter changeover during operation and pressure loss indicator as well as alarm PDIAH Testing device for checking the automatic start up of the auxiliary oil pump Control oil tank with auxiliary equipment Electric oil heaters Filling sieve and breather Degasifier plates Suction strainers Control oil pumps, motor-driven; one pump in operation, the other one acting as standby pump, starting automatically if control oil pressure drops Control oil filters, two in parallel each one sized for full flow, with transfer valve for filter changeover during operation and pressure loss alarm PDAH Testing device for checking the automatic start up of the standby pump Bladder-type pressure accumulators assuring that the compressor stator blades are closed properly in the event of control oil pump failure
509
5 lO
Process Plant
Machinery
Symbols for the above schematic diagrams Line symbols
Mechanical
P r o c e s s main G l a n d b a l a n c i n g line G l a n d l e a k a g e line 9,,,,,, L u b e oil s u p p l y ,,,-,,, J a c k i n g oil ,., -- L u b e oil r e t u r n line ,, = -, Oil mist :.'. ..... C o n t r o l oil s u p p l y -,,.. ,+,... C o n t r o l oil r e t u r n line ,,, ,,,, C o o l i n g w a t e r - - - - V e n t line 9 9 - Dust e v a c u a t i o n line 9 9 I n s t r u m e n t air -;- c- E l e c t r i c ~ e l e c t r o n i c s i g n a l Mechanical connection Valve
-r
and apparatus
symbols
Globe/ball/disc valve Gate valve V a l v e with c o n t r o l trim T h r e e - w a y valve GE~ N o n - r e t u r n v a l v e N o n - r e t u r n v a l v e with b o r e IZ] Suction s t r a i n e r Spring-loaded safety valve F l o w orifice Drainplug Double acting piston actuator Piston a c t u a t o r with c o u n t e r spring Diaphragm actuator Solenoid E l e c t r i c motoz Pressure accumulator Electric/pneumatic converter Electric/hydraulic converter ~ ) Low s i g n a l s e l e c t i n g r e l a y
+
m
symbols
~__+0,, F l e x i b l e c o u p l i n g Solid c o u p l i n g ~'-" J o u r n a l b e a r i n g =~ Thrust bearing, tilting-pad type [~] G e a r with thrust c o l l a r
..J"l_
Labyrinth gland chamber Instrument
symbols
M e a s u r e d v a r i a b l e s (first l e t t e r ) B Antisurge F Flow (3 Position H Hand L Level P Preasure PD P r e s s u r e d i f f e r e n c e T Temperature O u t p u t function (following l e t t e r ) A Alarm G S i g h t - g l a s s without m e a s u r e m e n t I Indication S Switch function E x t r e m e v a l u e s (last l e t t e r ) C Closed 0 Open H High L Low S i g n a l processing ~) C o m p r e s s o r d r i v e r s t a r t p e r m i s s i o n " ~ ' R ~ n i n g g e a r start p e r m i s s i o n 4) Compressor driver shutdown C ~ m p r e s s o r driver start up interlock Abbreviations
A
E x h a u s t to a t m o s p h e r e
D
Drain
V
Vent
Axial Flow Compressors
511
FIGURE 12-30 Layout drawing of a compressor, type AV, with driving motor and gear unit. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
Typical plant layout Figure 12-30 shows an example of a blast furnace blower plant. Figures 12-31 to 12-38 are photographs in various field situations.
512
Process Plant Machinery
FIGURE 12-31 Two steam-turbine-driven axial compressors delivering 330000 Nm3/h and 450000 Nm3/h of air at 5.3 bar to the blast-furnace plant of Hoogovens, ljmuiden (NL). Power input 29 000 KW and 40 700 kW respectively. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
FIGURE 12-32 One of the two motor-driven compressor~expander trains in the nitric acid plant of Norsk Hydro at HerCya (N). From left to right: axial tail-gas expander, 123 000 Nm3/h, 7700 kW; centrifugal nitrous gas HP compressor, 140000 Nm3/h, 4330 kW; axial nitrous gas LP compressor, 180000 Nm3/h, 12 150 kW. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
Axial Flow Compressors
513
FIGURE 12-33 Two axial water vapour compressors for a thermocompression/evaporation plant in the Austrian salt works Steinkogel. Capacity 81 t/h or 95 000 m3/h, power input 3520 kW each. The compressors are equipped with a special washing device, periodically injecting demineralized water in order to remove salt deposits which may otherwise accumulate on the critical internal parts within the flow-path. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
FIGURE 12-34 Turbocompressor train for the nitric acid plant of SASOL, Secunda (South Africa), on the test bed From left to right axial tail-gas expander 111 000 Nm3/h, 10950 kW; centrifugal nitrous gas compressor, 122 500 Nm3/h, axial air compressor, 139000 Nm3/h, 11890 kW condensing~extraction steam turbine 6500 kW, 6600 rev/min. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
514
Process Plant Machinery
FIGURE 12-35 one of the three identical two-casing axial compressor sets for the compression of 290000 Nm3/h of a hydrocarbon mixture from 1.42 to 39.2 bar, installed in the natural gas liquefaction plant (LNG) at Skikda (Algeria). Each steam-turbine-driven compressor train requires a power input of 78 MW. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
Completed air compressor module incorporating axial compressor A 71-10, 13-MW steam turbine and condenser. The direct driven compressor supplies 231 000 Nm3/h of catalyst regenerating air at 3.4 bar for a fluid catalytic cracking unit (FCCU) of Statoil at Mongstad, Norway. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland) FIGURE 12-36
Axial Flow Compressors
FIGURE 12-37 The word's first compressed-air energy storage (CAES) installation, operated by the Nordwestdeutsche Kraftwerke AG (NWK) at Huntorf (Germany). The compressor unit, consisting of an axial LP air compressor and a centrifugal HP compressor coupled through a gear to the motor/generator of the BBC gas turbine, pumps 300 000 Nm3/h at 46 to 72 bar into a salt cavern during the charging cycle and absorbs 60 MW. The gas turbine produces 290 MW during the peak-load period. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
FIGURE 12-38 Three of four identical axial compressor sets installed at the US Air Force Aeropropulsion System Test Facility of the Arnold Engineering Development Center, Tullahoma, Tennessee, USA. Each compressor delivers 352 000 Nma/h of air at 3 bar and absorbs 18000 kW. Another two 35-MW compressors of the same frame size can boost the air to 9.6 bar, but both LP and HP machines are operated at different pressure levels for aircraft engine tests at various simulated flight conditions. (Source: Sulzer Turbo Ltd., Ziirich, Switzerland)
515
This Page Intentionally Left Blank
Chapter 1 3 Propeller, Axial, and Centrifugal Fans* Fan applications in modern process plants range from handling fresh air to moving large volumes of corrosive and abrasive gas streams. This chapter deals with the fundamentals of selection and operation of fan systems. Industrial fans can be classified into three basic groups: propeller, axial and centrifugal. PROPELLER FANS
Propeller fans utilize long slender blades twisted in such a manner as to provide some angle of attack on the gas being moved. The blades are fixed to a hub, and the entire assembly rotates in a housing. The housing has little or no effect on controlling the gas flow. Typical applications for propeller fans are wall- and ceiling-mounted exhausters (Figure 13-1) and cooling tower and air-cooled heat exchangers. Pressures generally range from 0 to 1 inch of water column with efficiencies ranging from 10 percent to 35 percent. Tip speeds are often limited so as to minimize noise generation.
AXIAL FANS
Axial fans are basically propeller fans with shorter rigid blades assembled in a tubular housing to provide some degree of controlling and/or streamlining the gas flow. Figure 13-2 shows a small, simple, single-stage unit, while Figure 13-3 depicts a large, heavy-duty industrial version. This type of fan propels gas axially through its housing, which acts as an integral part of the ducting. Axial fans can be further subdivided into two types, namely, vane axial and tube axial. Because of the type of construction employed, these fans lend themselves easily to multistaging if higher pressures are required. Figure 13-4 illustrates a two-stage axial fan. Vane axial fans have vanes installed before and/or after the rotor to direct and streamline the gases for better fan action. The vanes are installed on the stator housing and employ some kind of control mechanism for varying the flow pattern. Vane axial fans are used primarily in clean gas service, handling pressures generally up to 20 inches of water column (single-staging), but higher pressures are available. Multistaged vane axial fans can achieve efficiencies in excess of 85 percent and are generally considerably more efficient than tube axial fans. * Source: Garden City Fan and Blower Company, Niles, MI (except as noted). Adapted by permission. 517
518
Process Plant Machinery
FIGURE 13-1 Wall-mounted exhaust fan (propeller-type). (Source: A CME Engineering & Manufacturing Co., Muskogee, "OK.)
FIGURE 13-2
gee, OK.)
Small axial fan. (Source: ACME Engineering & Manufacturing Co., Musko-
Propeller, Axial, and Centrifugal Fans
t
FIGURE 13-3 Heavy-duty single-stage axial fan. 1 - Access door; 2 - diffuser; 3 - external pitch control lever; 4 - stationary blades; 5 - variable pitch rotating blades; 6 - removable upper fan housing; 7 - main bearing assembly; 8 - coupling; 9 - inlet box; 1 0 - motor; 11 - blade pitch control mechanism; 12 - rotor assembly; 13 - shaft tube; 14 - drive shaft. (Source: C E M A X Fans, Boston, MA.)
FIGURE 13-4 Heavy-duty two-stage axial fan. 1 - Access door; 2 - diffuser; 3 - removable stationary blade; 4 - bolted blade retainers; 5 - removable variable pitch rotating blades; 6 - removable upper fan housing; 7 - coupling; 8 - external pitch control lever; 9 - blade pitch control mechanism; 1 0 - rotor assembly; 11 - main bearing assembly; 12 - drive shaft; 13 - shaft tube; 1 4 - shaft seal. (Source: C E M A X Fans, Boston, MA.) Tube axial fans employ the same basic construction as vane axial fans except for the guide vanes and are generally less efficient than vane axial fans. Recently, variable-pitch rotor blade construction has been offered as a standard package for both tube and vane axial fans. Typical performance characteristics of axial flow fans are shown in Figure 13-5. Note the pronounced "dip," or stall range, in both the static pressure and brake
519
520
Process Plant Machinery
4W
~v
uu
~u
loo
120
PEFtCE:]~T OF WIDE OPEN VOLLlUE
FIGURE 13-5
Axial flow fan characteristics
horsepower (BHP) curves. This represents the unstable range in which surge phenomena occur and where fan operation should be avoided. It also explains why axial fans operating in parallel are difficult to start up. Although the absorbed horsepower is initially high, its subsequent decline makes it a non-overloading characteristic.
CENTRIFUGAL FANS Centrifugal fans (Figure 13-6) are prevalent in virtually all of the industry. These fans are applied in building and equipment ventilation, hot-gas recirculation, dusthandling systems, and furnace and boiler forced draft/induced draft services. The basic difference between centrifugal fans and the types previously mentioned is the fact that gas enters the impeller axially and leaves the rotating element radially. The casing is designed and installed around the rotating element to direct the discharge gas in a manner suited for the particular application. Centrifugal fans operate with typical efficiencies in excess of 80 percent and an achievable limit of 91 percent with airfoil blade construction. Discharge pressures can be as high as 90 inches of water column or higher and cover a wide operating range. Centrifugal fans can be categorized by three basic construction styles depending on blade orientation with respect to impeller rotation. These styles are radial blade, backward-inclined blade, and forward-inclined blade. Subtle variations of each style are available.
Radial Blade As the name implies, the blades on this type of impeller are radially oriented. This type of impeller construction can be further subdivided into straight radial blade and radial-tip blade styles. The merits of each of these different blades can be observed in Figures 13-7 through 13-10. The straight radial bladed fan illustrated in Figure 13-7 has a best efficiency of approximately 72 percent. The fan has an
Propeller, Axial, and Centrifugal Fans
FIGURE 13-6 Centrifugalfan for typical process plant. (Source: A CME Engineering & Manufacturing Co., Muskogee, OK.)
i
150 "j 140 "1 130 - I
120 --~r-
--
i
70 - [ f
i- Z 0
2O
J
J I
40
SO
P ~ O E ~ OF WiDE OPEN VOLUUE I[
FIGURE 13-7
Radial blade fan characteristics
521
522
ProcessPlant Machinery
mmmmmmmmmm~qmmmmmmmmmm m ~ ~ m m m i ~ ~ ~ m m ~ ~ ~ i ~ a ~ - , m ~ m m mmmm. mmmmn~mm--r~Immm m ~ ~ ~ ~ ~ m m - ~ ~ ~ ~ m m ~ ~ ~ ~ ~ m ~ ~ 1 ~ ~ m m i ~ m ~ ~ m m ~ ~ . ~ ~ m m ~ ~ ~ ~ ~ ~ . ~ ~ ~ ~ m m m m m m m ~ ~ m m m m m 0
mmmmmmmmmmmmmmmmmm, 20
40
60
mmm
BO
100
120
PERCENT OF V~DE OPENVOLUUE
FIGURE 13-8
Radial-tip blade fan performance characteristics
overloading horsepower characteristic, but it features stable operation over the entire performance range. Because of low specific speed, straight radial bladed fans have relatively good resistance to abrasion from entrained particles. The radial-tip fan of Figure 13-8 minimizes gas turbulence by curving the blades. It achieves higher efficiencies, with 78 percent to 83 percent being not uncommon. Fan horsepower is nonoverloading in the high flow range. This fan also has low specific speed and medium resistance to abrasion. Figure 13-9 illustrates an open radial bladed fan with a geometry that readily allows the addition of simple wear plates for highly abrasive applications. It is
160 1.4,0 I,}0 720 110
r
,oo
90
,,
rr
~TA,~C ~'RESSU~E
T 1
/-
80
70
1~
~
f
~
&---"-'
L_ rr,,nc
\
BHP
-
~o . . . . ~ 0 0
20
40
60
BO
! O0
120
PERCENT OF ~ D E OPEN VOLUUE
FIGURE 13-9
Open radial blade fan characteristics
Propeller, Axial, and Centrifugal Fans lowest in efficiency, with 65 percent being typical. This fan has an overloading power characteristic. In summary, radial blade fans are commonly used for induced draft service and are especially suited for contaminated gas streams where airborne particles and fouling could be a problem. Backward-Inclined
Blades
The fan blades on this style of rotor are oriented backwards with respect to the direction of rotation. Three variations of backward-inclined blades are available. Figures 13-10 through 13-12 illustrate both geometry and performance of these variations: backward-inclined fiat plates, curved plates, and airfoil blades, with efficiencies increasing in that order. Typical efficiencies for flat plate blades range from 77 percent to 80 percent and 84 percent to 91 percent for airfoil blades. The flat and curved plate blades can handle fine airborne particles that would normally damage the airfoil design. These fans are commonly used in forced draft services. Modification of the backward-inclined flat plate profile of Figure 13-10 reduces the unstable range to moderate instability in the backward-curved blade (Figure 13-11) and virtually no instability in the backward-inclined airfoil blade (Figure 13-12). Efficiencies show slight increases in the order mentioned; however, the level of tolerance for handling abrasive gas streams decreases as one progresses from backward-inclined to airfoil geometries. All of the backward-inclined blade fans exhibit non-overloading horsepower characteristics. Forward-Inclined
Blades
These fans are styled such that the blades lean forward with respect to the direction of rotation. The two typical variations are the fiat plate and curved plate. Figure 13-13 shows geometry and performance of the forward-inclined curved blade style. This
L_2_
t
,oo
[
_! - ~ - I. . . . i :
7o
I
T--TL
' ."
.... rzj-I
i-T---T--X-W-,*---~ .... ~ - - - t - - - 1 ...... ~ ~ - * - - . . . . . . . +-. . . . t-i----I. . . . . . . .
- r - - - " - - ' ~ - ~ -
!
I-
i,
i
-
'.........
! '
11--T"~-F-'~
- - -
i
'T -
.....
[
i
~ - ~ - - ~ - .,..---%....;~--H---~--- , ~ + ~ ....; --,~--
,o..h---~.+----l.-..---t-:-.--t-.._~-__~,. 3o
-1----r
...... . . . .
_t .~ t+_ I + -i o,~
s
,o~-
0
.... ,",++
I
. I
20
.
.
~ .
.
. . . . TI - T -I ~ 40
.J ' I
1
.....
...... i--'-,:-k--.:,----4- . . . . l~ - - ~ M - - ~
. . . . i ..... ...... ....
+ 't-'.~A_~c O'mf=~c~J " ~ I '
_ ,I
60
....
-+--
...~I
--. ......
+
.
,
I
80
..l
I
~.
l 9
. . . . . . . T ....
100
I
-
120
PERa[~ OF WIDE OPEN VOUJME
FIGURE 13-10
Backward-inclined blade fan characteristics
523
524
Process Plant Machinery
E l
,
i
!
_.~
~J~
1
....
I .~
r
l
..__
~'i--~.~-~-~____ J
liil
A
x,,,L
xx..r
i
40
",, x,,,~ \
ii
r L BHP
~.,
~'A1 IC EFFICll )ie't'
I1 -i I 1 20
\i
...
I
0
t
1N
,
.1:_
-4 . . . . . I
q.,, N
[
~-~
-~- .......
[
i ......
,-9
i
I .
.t
I
i
.
i
$ TATIC ES~JRII
-~j
60
80
100
120
POtCOIT OFWIDEOPENVOLUME
FIGURE
160
. . . . . . . . . . .
,~o-t----t.
13-11
I
Backward-curved blade fan characteristics
.
.
.
.
.
...... -i,-..... I/-t- ....... ~.... -i ....... .-!---I......
,oo:~-~--:
i .....
: .....
r--
........ ~ .......... ~,- -~ .
te
i
~ ---
.
.
....
~
..... ~,- ~.-+--~
.
i .... I
~
,
.t---- - i ....... r ...... -l--Jr
o~-~
lo
...... 9
,o
.... ~r~
_
,_, ........
,t-- .... -'r,, - - + \ x ~
l-~r,Ti;4 ---
-- -~. .... ~,...... l- --~,~d~--t
4 I ~,:~-~
' .... b . . . . . . . . . . . . . .
'-L ..... ....
~ ......
[-----1
--I----~---1
-V~,i..-~r ....
.... ~t r .....
~ ....
,~.:~TO,,,o~:,~-o.r 0 '>< i
A
B
Time-0
Time- 1
Time-2
Time-3
Time-4
Upper half
4A+4B
3A ~2B
3A+0B
2A+08
2A+OB
Lower haft
4A+4B
4A+6B
5A+BB
5A+88
6A+BB
FIGURE 17-15 Principles of the conventional classification method. (Source: Dorr-Oliver Inc, Millford CT.)
Improved Classification
Method
By displacement method
1
Time..0 Overltow
i"i"l
Time.,1 1A
ITi
~
Time-2
ITI
Time-3
2A
3A
Time.4 4A
Upper half
4A§
5A.18
6A.0B
7A.0B
8At0B
Lower half
4A.4B
3A,,6B
2AtSB
tA+SB
0A.SB
FIGURE 17-16
Millford CT.)
Principles of the improved classification method. (Source: Dorr-Oliver Inc,
Separators 647 The fine particles in the overflow can be thickened in another centrifuge and the clear liquid fraction can be recycled back to serve as the wash stream. Obviously, the upflow rate of the wash liquid is critical and specific to the situation. If its velocity is too high, all of the spheres may be lost and if the velocity is too low, then some contaminating small spheres may remain with the large spheres. One restriction to this operation is that the size range of the small particles cannot overlap the size range of the large particles. In the overlapping case, a sharp separation cannot be achieved. Merco centrifuges have washing capabilities which provide a unique feature allowing the operation of classification of particles simply by following the principles illustrated. APPLICATIONS
Three kinds of processes, i.e. classification of kaolin clay particles, washing of terephthalic acid crystals, and dewaxing of lube oil, are employed for the purpose of discussing the application of the disk nozzle centrifuge in the purification of solids mixtures occurring in the chemical industry. During the processing of kaolin clay, the raw material is mined and slurried and the coarse sand and grit is removed. Several stages of centrifugation are employed in order to separate the slurry into various fractions for other uses. Such equipment includes decanter centrifuges and batteries of hydrocyclones. High-speed disk nozzle centrifuges are used in several positions in the flowsheet but the most critical station is for "fractionation." The slurry is a mixture of clay solids suspended in water and the solid material is all of the same density and character except for size. The most valuable clay fraction is less than 2 microns but larger than 0.5 microns in size and is used to make paper whiter and improve the appearance of print. Here is where we "cut out" the slimes and leave the bulk of the supply as product. A typical feed would have 30-40% DS and the"underflow" product would be 5 0 - 6 0 % DS and the "overflow" product would be 15-20% DS depending upon the product requirement. The disk nozzle centrifuge would typically operate at high speed and with a large number of large size rotor nozzles so as to cope with the flow requirements. The feed rate to the centrifuge and the underflow compositon (% solids) are controlled by valves so as to render the desired "cut" (the overflow). Operation in this way is quite different than when a centrifuge is in a "clarification" mode. Both fractions are opaquely white and the sharpness of the "cut" requires that the particle size distribution of each fraction be determined. The analyses are done using special instruments. By using the elutriation wash principle, as described earlier, it is possible to remove the slimes more completely and to simultaneously recover more of the valuable material. The action takes place exactly as explained. The wash is introduced into the centrifuge rotor at the bottom and it enters the slurry bed at the "bottom" (the inboard periphery near the discharge nozzles). The water rises through the suspended bed carrying slimes with it and the flow passes on through the separating disks and discharges as "overflow". Accordingly, the underflow has less slimes remaining in it and the overflow has more. There is some dilution of the overflow due to the additional water. When this scheme is practiced, a series of runs must first be conducted at various wash water addition rates. The results will show how much benefit each increment of water provides. One can then calculate the optimum combination of feed and wash rates to perform the purification. According to the test results, normal classification by differential settling of kaolin particles gives 75% recovery of the
648
Process Plant Machinery
most valued fraction, whereas classification enhanced by elutriating wash results in 85% recovery. The processing of terephthalic acid affords a rather novel example of the benefit of displacement washing when using a disk nozzle centrifuge. In this case, the product (PTA crystals) is washed while it is suspended in a hot corrosive fluid under elevated pressure and temperature conditions. Terephthalic acid is made in many countries around the world and by a variety of processes. Some of its uses are fiber, film, and plastic bottles. In most processes centrifuges are employed to separate the solid crystals (less than 200 microns in size) from the mother liquor and a counter-current washing scheme is used to assure the production of purified crystals. The solids content of the feed may be of the order of 35% DS and the underflow may be exactly the same or at a higher concentration. The large size of the crystals (peaking around 70 microns and none smaller than 20 microns) makes it possible to operate with high upflows of the wash fluid. Most of the Merco centrifuges operate at 10 bar, or less, pressure but some models operate at 20 bar. Higher pressures and temperatures improve the performance by reducing the fluid viscosity. High-speed disk nozzle centrifuges are used in series where the displacement wash efficiency can be over 90% per stage if high wash ratios are used, such as W / U of 2.2. These particular centrifuges have outstanding performance records for on-stream time and have required very little mechanical maintenance. A third example involves the processing of petroleum wax which is removed from lube oil by chilling the feed stock together with the introduction of certain solvents which enhance crystal formation. In this process the conventional arrangement is to use rotary vacuum filters in series in order to completely dewax the oil. The procedure is practiced around the world in similar fashion but on differing flow scales. Merco disk nozzle centrifuges are uniquely able to perform in this process even though the density difference between the wax crystals and some mixtures of solvents and oil are very small as well as having the handicap of higher-thannormal viscosities. The centrifuges are completely sealed and may be pressurized, as required. In this example, the percentage of wax crystals is less than in the previous cases and this value will vary with the temperature. Even a small rise in temperature will liquefy some w a x - there is no fixed melting point. Both the wax and oil vary according to the source of the crude oil and specific testing has to be done in order to define the performance in each case. Fortunately, this can be done easily. It is also important not to have emissions. This is for environmental pollution reasons as well as the hazard of fire. In summary, several criteria must be met in order to use a disk nozzle centrifuge for the purification of a solids-liquid slurry, as follows: 1. A continuous supply of a large volume stream of a solids-liquid mixture. 2. An impurity being present that can be separated away by virtue of a differential settling rate to a solids fraction (includes classification and washing). 3. The solids to be small (below 1 micron up to 200 microns) and difficult to settle (otherwise a simple gravity device would suffice). 4. The product to have a substantial value. 5. A complex set of operating conditions can be overcome by virtue of having diverse experience already (corrosion, temperature, pressure, solvents, hazards). 6. High solids concentrations can be reached and high wash rates can be used as well as using multi-stage counter-current systems.
Separators 649 Dilution and displacement are the two washing modes used in process industries. Displacement washing is the more effective of the two, as with substantially less wash, more liquor can be displaced. Disk nozzle centrifuges with internal recirculation are uniquely equipped for displacement washing, because the wash is introduced into a preconcentrated underflow slurry, of which most of the mother liquor had been previously displaced within the separation chamber. Using equal amounts of wash versus feed liquor, displacement washing achieves a 70% removal while dilution washing is limited to 50%. Increasing the wash liquor to 2:1 improves the washing efficiency to 90% versus 67% for dilution washing. Washing of starch to remove protein, washing of TPA to remove soluble constituents and the removal of wax fractions from petroleum are just some of the more common and proven applications. The same principle of upflow used in displacement washing is used for classification purposes to classify very fine from coarse particles. A typical application is in kaolin where 0.5 micron or smaller particles are separated from 2 micron material. Centrifuges have the added advantage of being totally enclosed, contain a minimum of (process liquid) inventory, operate in a continuous mode, can be easily temperature controlled and are available in corrosion resistant materials.
This Page Intentionally Left Blank
Chapter 1 8 Internal Mixers: Single- and Twin-Screw Extruders* INTERNAL MIXERS Internal (Banburyt-type) mixers have found widespread application in the rubber industry, where high-horsepower mixing is required for masticating and compounding; the technology has been successfully applied to plastics and chemicals for high- and low-viscosity systems:
Rubber Applications
Plastic Compounds
Other Applications
Tires, tubes
Polyvinylchloride (PVC)s (flexible, semirigid)
Adhesives, sealants Carbon electrodes Ceramics Chewing gum
Packing, sealing, roofing
PVC scrap reclaim ABS (molding) Polyethylene, polypropylene, PVC, ABS color concentrates Phenolic
Flooring (sheet)
Thermoplastic/rubber blends
Molded articles, profiles Hoses, gaskets, seals Shoe soles, heels Foam, sponge rubber
Dewatering, devolatilizing Pharmaceuticals
The common goal is to mix solid and/or liquid additives into a rubber or plastictype matrix. Additives (agglomerated particles or droplets bound by surface tension) must be separated, reduced in size, and uniformly distributed within the matrix: Two types of mixing phenomena are involved: extensive mixing and intensive mixing. Extensive (also known as distributive) mixing is responsible for spatial distribution of the individual particles within the polymeric matrix. Intensive (also known as dispersive) mixing is responsible for separating and reducing the particle size of the additives.
Principle of Operation Internal mixers are designed to provide intensive and extensive mixing. Intensive mixing occurs in the narrow gap formed between the rotor tip and the mixing chamber wall. The mixture is repeatedly passed through this high shear field where fluid mechanical stresses separate and rupture agglomerates; dispersive forces are * Source: Werner & Pfleiderer Corporation, Ramsey, NJ. Adapted with permission. t Banbury is a registered trademark identifying internal batch mixers made by Farrel Company and its predecessor companies since 1916. 651
652
Process Plant Machinery
similar to those in a two-roll mill. Extensive mixing takes place between the rotors: the mixture is circulated from side to side and from one end of the mixer to the other after passing through the shear zone. Energy dissipation from intensive mixing results in heating of the mixture. This heat is removed through the walls of the mixing chamber, the rotor bodies, and other contact parts (ram, discharge door, etc.) through cooling channels. The effective heat transfer of internal mixers can be the limiting factor to intensive work, since discharge temperatures cannot exceed the critical temperature of the mixture, which is determined, for example, by thermal breakdown of organic phase, onset of undesirable reactions (e.g., cross-linking), or the decrease of continuous-phase viscosity to a point where dispersion cannot proceed. Poor heat transfer will cause these temperature limits to be reached before dispersion is complete.
Design Features of Internal Mixers Optimum dispersion is achieved through proper selection of machine type and process parameters. Internal mixers are available today with intermeshing and tangential (nonintermeshing) rotors. Rotor design is a critical factor in mixer performance, and some manufacturers offer various rotor configurations. (This topic is covered in detail later). Figure 18-1 shows the major components of internal mixers. A completely enclosed mixing chamber houses the spiral-shaped rotors, which rotate in opposite directions and at the same speed (intermeshing design) or at different speeds (tangential design) to keep the material circulating. The gap between the rotor tips and the chamber wall produces intensive shearing of the mixture. A hopper allows for loading ingredients, and an air-operated ram in the feeding neck confines the batch within the mixing chamber. A discharge door allows for quick and efficient unloading of the mixture at the end of the mixing cycle. The mixing chamber, rotors, and discharge door are all temperature controlled with steam and/or water.
Mixing Chamber The mixing chamber body is usually of two-piece construction, split vertically. This allows for removal on-site without dismantling the entire mixer. The internal body halves are lined with wear-resistant materials to maintain rotor tip-to-wall clearances; the exterior is heavily reinforced for mechanical loading. Temperature control of the mixing chamber is accomplished with steam or circulating water. The body halves are fitted with one of three types of chamber sides: cored, spray-type, or drilled. Cored sides have passages arranged in a serpentine pattern running along the length of the sides of the chamber. These passages are usually formed in the casting, through which water or steam is circulated. Spray-type sides use nozzles to spray water onto the sides for cooling effect; flood-type design is also available, without nozzles. Drilled sides are the most common type of temperature control for the mixing chamber. Holes are drilled laterally in the body halves and provide a serpentine path for the flow of water. These holes are smaller and greater in number than the cored passages, as well as being closer in proximity to the chamber wall.
End Frames The ends of the mixing chamber provide support for the mixer body. End frames carry the rotor bearing assemblies and dust-stop seals.
Internal Mixers: Single- and Twin-Screw Extruders
".4Pll'/
:
i\
i
;
I
\\ \
J,
J
I I
i
\
/
/
\
FIGURE 18-1 Cross-sectional view of internal mixer with intermeshing rotors. (Source: Werner & Pfleiderer Corporation, Ramsey, NJ.)
653
654
Process Plant Machinery
Self-aligning roller beatings are standard on most internal mixers. Double-acting axial thrust bearings can be used on large internal mixers to increase lifetime and efficiency of the rotor seals. Dust stops are used to seal dust (carbon black, pigments, etc., used in the mix) within the mixing chamber. There are several designs available; one type relies on a hydraulically actuated yoke to apply pressure on the sealing rings. Self-sealing dust stops use the mechanical pressure of the mixture to maintain seals. Powders or dust that pass into the dust stops are flushed with process oil.
Discharge Doors The discharge door of internal mixers is designed to provide quick and efficient dumping of the mixture. The door top must be hard-surfaced to withstand the mixing environment; some designs include a removable/replacement door top section. Most discharge doors are provided with passages for heating or cooling media; the door is usually cooled to prevent the mixture from sticking. Two types are available: drop door and sliding door. The drop door type is pivoted on a shaft running through bearings located in the end frames. The door swings downward (135 ~ to 180 ~) and away from the mixing chamber, providing a clear path for the material discharge. The sliding door is mounted on an air-operated cylinder. Guides are necessary to provide clearances for door operation. Also available are mixing chambers that tilt over (up to 140 ~) to discharge the mixture through the top opening. Tilting mechanisms are electrically or hydraulically driven.
Feed Hopper The feeding of materials into an internal mixer can actually take longer than the mixing cycle. Thus, efficient designs are utilized on the hopper assembly of internal mixers to facilitate quick charging of ingredients. The feed opening itself must be large to allow venting of air (and dust) as the batch is charged. The feed opening and throat on larger mixers are sized to accommodate rubber bales intact. Air-operated doors are provided in the hopper assembly to allow for loading of material into the mixer. Openings on the sides or in the rear of the hopper for charging of fillers, accelerators, curatives, and other dry powder components are normally connected to weigh hoppers or other means of feeding.
Ram Cylinder Assembly A ram is used to confine the batch within the mixing chamber. Air pressure (10 to 120 psi) is applied to the cylinder, forcing the ram down into the mixing chamber. The bottom of the ram is usually shaped to conform to the gap between the rotors (V-bottom). The ram is fitted with a height indicator, used to gauge the state of the mix. Cavity cooling is used for temperature control of the ram bottom. The ram cylinder assembly is usually air-operated, but it can also be a hydraulic device. Ram pressure can have a significant impact on dispersion quality, and it is sometimes used as an operating parameter.
Drive Train There are three mixer drive arrangements available: "standard," semi-unidrive, and unidrive. Standard drive trains use reduction gears mounted on the mixer base to drive the rotors; one rotor shaft is longer than the other, functioning as pinion shaft and reduction gear. Semi-unidrive systems use a separate reduction gearbox prior to the rotor shaft. One rotor shaft is longer than the other, carrying the pinion gear. Unidrive systems have speed reduction and dual output shafts within a
Internal Mixers: Single- and Twin-Screw Extruders single gearbox. Rotor shafts of equal length are coupled to the pinion shafts from the gearbox. Rotor speed directly influences mixing quality, mixing time, and batch temperature. Optimum rotor speeds are chosen to process materials at their highest acceptable temperature within the shortest cycle time. A single-speed drive limits the number of formulations that can be processed optimally. Two-speed motors increase internal mixer flexibility. Several other alternatives are also available: 9 constant speed motor with integral (or separate) two-or four-speed gearbox 9 variable-speed DC or variable frequency AC motor (ultimate flexibility)
Bed Plate The bed plate is the base frame that anchors the mixing chamber (and possibly gear-box). It is strengthened to withstand torque transfer and vibration of the rotors, as well as to evenly distribute the load of the mixer onto a foundation.
Auxiliary Systems Operation of internal mixers requires several dedicated systems. The tempering system supplies constant-temperature fluid to heat or cool the various mixer components. Steam/water, pressurized water, or heat-transfer oils can be used. Several separate circuits may be needed for efficient operation (e.g., ram and mixing chamber at one temperature, discharge door at another temperature, and rotors at yet a third temperature). The lubricating oil system ensures adequate supply of lube oil to critical components in the drive train, rotor bearings, dust stops, etc. The process oil injection system injects oil into the mixing chamber. The injection nozzle should be a self-sealing type to prevent fouling from mixture. The hydraulic or pneumatic system operates discharge door (or tilting mechanism), ram cylinder, feed chute door, etc. It is controlled by means of solenoids. The temperature sensor, strategically located, indicates batch temperature. It may be located in the feeding ram, in the discharge door, or through the end frames. Sensors should have intimate contact with the mixture to provide accurate readings.
Instrumentation and Controls The degree of sophistication of internal mixer control systems can vary. They can 1. 2. 3. 4.
mix batch until desired temperature is reached mix batch for predetermined time period mix batch until predetermined energy is consumed use various combinations of the above
Efficient operation of internal mixers is attained with automation of the mixing process. Increased productivity and consistency in product quality can be realized with computerized control of the mixing line from batch weighing through downstream processing. Manual Control. Standard internal mixer control systems provide interlocks and data acquisition for manual operation. Interlocks are installed on ram cylinders, feed hoppers, and discharge doors for operator safety and on drive components for overload protection. Operating data are usually recorded on strip-charts housed in a control panel for production monitoring of discharge temperature, power
655
656
Process Plant Machinery
consumption, and cycle times. Stop/start push buttons are provided for drive motor and auxiliary equipment. Automated Mixer Control Systems. Application of a process computer system to an internal mixer significantly improves batch quality as well as quality consistency from batch to batch. Flow of raw materials, mixing control, downstream equipment, and production planning can all be integrated into a supervisory computer system. Optimum mixer control is achieved with logic controls and/or combinations of mixing time, energy input, rotor speed, stock temperature, and chemical reaction parameters. Adaptive process control systems allow the mixing process to follow predefined energy and temperature curves that are stored in memory with each formulation. Rotors
Mixing is achieved in internal mixers with the rotors, rotating toward each other at the same speed (intermeshing design) or at different speeds (tangential design). The rotors are designed to interact with each other via the rotor blades, called wings. Short wings and long wings are used in combination on each rotor. Mixers are available with either two-wing or four-wing rotors. Two-wing rotors have one short and one long wing, while four-wing rotors have two short and two long wings. The rotors are arranged in the mixer such that the long wing of one rotor interacts with the short wing of the other rotor (Figure 18-2). The edge of the blade is called the wing tip, which forms the shearing gap between the rotor blade and the chamber wall. The rotors are temperature controlled with steam and/or water, which flows through the center of each shaft. Spray-type cooling or forced circulation is available. Rotors and shafts can be manufactured as a one-piece steel casting or as two pieces: forged steel rotor shafts with rotor bodies shrunk onto the shafts. The entire rotor bodies are hard-surfaced for wear protection or are chrome plated with hardfacing only on the tips.
FIGURE 18-2 Plan view of tangential four-wing rotors, rotor bearings, and dust stops. (Source: Werner & Pfleiderer Corporation, Ramsey, NJ. )
Internal Mixers: Single- and Twin-Screw Extruders
Rotor Design Intensive mixing occurs where the material is compressed between the rotor wing tip and the mixing chamber wall. The width of the tip, the clearance between the wing tip and the chamber wall, and the leading/trailing angle affect dispersion. Wing length and helix angle influence the distribute mixing from rotor to rotor and from one end of the mixer to the other.
Tangential Rotors Conventional internal mixer designs use tangential rotors with two or four wings. The rotor diameter is equivalent to the center distance between the rotors. Tangential rotor mixers are mainly used for large-volume mixing, as in the tire industry. Two-wing rotors are characterized by two flow regions due to different radial clearances between the rotors. Mixing intensity is directly influenced by ram pressure with a two-wing rotor system. Four-wing rotors provide constant radial clearances between the rotors. Total wing length is greater than the two-wing system, producing higher specific energy input and better homogeneity. The torque capacity of internal mixers supplied with four-wing rotors is approximately 30 percent greater than with the two-wing rotor system. Modified four-wing rotor designs have been developed with enhanced longitudinal (distributive) mixing.
Intermeshing Rotors Manufacturers of internal mixers have developed intermeshing rotor systems in response to quality and productivity requirements of compounders. The diameter of intermeshing rotors is greater than the center distance between the rotors (Figure 18-3). A calendar effect is created by the intermeshing geometry of the wings, resulting in improved dispersion. Intermeshing rotor systems are capable of higher rates of energy input and better heat transfer than tangential designs. The number of mixing steps, as well as mixing time for each step, may be reduced by changing from tangential to intermeshing rotors. Internal mixers with intermeshing rotors are used mainly for high-quality mixing.
Operation of Internal Mixers Several mixing techniques are commonly practiced using internal mixers: singlestage, masterbatch, and multistage mixing. Single-stage mixing is used for materials that can tolerate relatively long mixing time at low rotor speed; temperature rise is the limiting factor. All ingredients can be charged at the same time or added sequentially while mixing. The sequence of addition of plasticizers and oils is critical to dispersion quality. Masterbatching is an implied two-stage mixing process. Viscous components and fillers are mixed first. This stage can tolerate higher rotor speed and material temperature. Thus, short mix cycles produce complete dispersion. The batch is discharged and allowed to cool. A fraction of the first stage (masterbatch) is then loaded with the balance of ingredients that make up the total formulation. The predispersed masterbatch mixes efficiently and can produce higher quality dispersions in less total mixing time than a comparable single-stage mixing process. Multistage mixing can include several masterbatching steps, a remilling stage to disperse the masterbatch, and final mixing.
657
658
Process Plant Machinery
FIGURE 18-3 Tangential (top) and intermeshing (bottom) rotor geometries. (Source: Werner & Pfleiderer Corporation, Ramsey, NJ.)
Operating Parameters Optimization of internal mixers requires knowledge of how operational parameters affect mixing quality. Generalized statements can be made as to the influence of fill factor, mixing sequence, ram pressure, rotor speed, mixing time, and temperature control. Fill factor is an indication of the working volume of the mixer. Chamber volume is used to specify internal mixing capacity (in liters or cubic inches); empty volume is measured with the rotors installed. The working volume is the space occupied by the mixture (function of batch weight, specific gravity). The ratio of working to empty volume gives the fill factor. Internal mixers are available with chamber volumes up to 650 liters (1), which can handle a 560-kilogram (kg) batch; laboratory-scale internal mixers used for research and development have net volumes as low as 0.41, requiring less than 0.3 kg per batch. Fill factors vary for different mixer geometries (two-wing versus four-wing; intermeshing versus tangential) and different mixing tasks. Intermeshing rotor mixers generally use lower fill factors than tangential rotor systems, but they can achieve the same level of dispersion in a shorter mixing time. A compound that runs on a tangential mixer at 70 to 75 percent fill factor (four-wing rotor) would run between 60 and 65 percent fill factor on an intermeshing mixer.
Internal Mixers: Single- and Twin-Screw Extruders
Fill factor directly influences dispersion quality, specific energy input, and discharge temperature by providing empty volume for the mixture to circulate within. Mixing sequence has a significant effect on dispersion quality, specific energy input, and discharge temperature when large amounts of oil are being processed. Oil or plasticizers are typically added later in the mixing cycle, after fillers have been dispersed. Ram pressure is used as a process variable to influence the specific energy input and discharge temperature. Tangential rotor mixers are more sensitive to ram pressure than intermeshing rotor mixers. Increased ram pressure can shorten the mixing cycle by compressing the batch within the chamber to intensify mixing. Discharge temperature can be reduced by decreasing ram pressure. Rotor speed is used to control the rate of specific energy input. Higher rotor speeds can reduce mixing time by dissipating more energy in a shorter time period. Rotor speed and ram pressure can be independently varied to achieve a target energy input, dispersion quality, or material temperature. Internal mixers supplied with a variable-speed motor have more flexibility in optimizing rotor speed for a given formulation than a single- or two-speed mixer. Large internal mixers operate at rotor speeds of 10 to 60 RPM; laboratory-scale internal mixers operate at speeds up to 110 RPM or higher. Mixing time in conjunction with fill factor (batch weight) determines the throughput capacity of the mixer. Decreasing the mixing time (by increasing rotor speed or ram pressure) then increases mixer output. Mixing times vary widely depending on formulation and quality of mix. For example, single-stage mixing of one particular rubber formulation and oil, which takes eight minutes on a tangential mixer, can be processed in only five minutes on an intermeshing mixer. A typical tire formulation running on a tangential rotor system takes about three minutes for the masterbatch stage, three minutes for the remilling stage, and two minutes for final mixing. Temperature control of mixer components (chamber body, rotors, ram, and discharge door) has several effects: adherance of material to rotors (rotor temperature), discharge temperature (cooling effect through chamber body), and efficiency of discharge (discharge door temperature). The mixer is usually started up hot to prevent slippage, and then cooling is applied when the material is in a fluxed state. Accurate temperature control of each part of the internal mixer helps maintain batch-to-batch uniformity.
Downstream Equipment An internal mixer usually discharges directly into some type of shaping/forming equipment or onto conveyors. Downstream equipment is sized to provide a continuous process from the batch mixer; two mixers can also be arranged with alternating discharge.
Mixing Mill Roll mills are used for cooling and shaping as well as for after-homogenizing and mixing. Cross-linking chemicals that cannot be added in the internal mixer due to temperature limitations in single-stage mixing can be added in the mixing mill. Mixing mills are built with fixed friction ratios (constant speed on rolls), or variable friction can be achieved with variable-speed control of each roll. Adjustment of the roll gap can be carded out under load with hydraulics.
659
660
Process Plant Machinery
Extruder Single-screw extruders are used to produce pellets or sheet from the internal mixer. Pelletizing extruders are fitted with screens or strainers and die-face pelletizers or underwater pelletizers. Force-feeding devices (screws or rams) can also be installed. Sheet stock is produced from extruders equipped with a roller-die.
Maintenance of Internal Mixers Routine maintenance of internal mixers requires attention to the various subsystems responsible for smooth operation (e.g., hydraulic, pneumatic, and temperaturecontrol systems). Mixer manufacturers provide recommendations for preventive maintenance on rotor bearings and seals, lubricating oil changes, gearbox overhauls, etc. High on-stream factors are achieved when maintenance records are kept and factory recommendations are followed. Major overhauls of internal mixers are required when the product specifications (dispersion quality, discharge temperature, etc.) are no longer acceptable. Abrasive wear on rotors, chamber body, and ram becomes evident in mixer performance: dispersion quality cannot be maintained as clearances increase between rotor tip and chamber wall; discharge temperature increases as a result of increased clearances and subsequent reduction of heat transfer. Manufacturers of internal mixers can provide a field inspection service to periodically document wear. Critical components (rotors, bearings, dust stops, etc.) are usually kept in stock for emergency delivery. Rebuilding of internal mixers is a service performed by the manufacturers. Worn rotors and chamber bodies can be rebuilt to factory specifications. New duststop parts, door tops, and end frames are installed. Older mixer designs can be converted from cored or spray-type sides to drilled sides, from two-wing rotors to four-wing rotors, from standard gear to unidrive, etc.
SINGLE- AND TWIN-SCREW EXTRUDERS Screw extruders were designed as continuous mixers for dispersing additives into a molten polymer. Intensive or extensive mixing takes place in the extruder as a function of screw geometry and operating conditions. Mixing time in continuous mixers cannot be arbitrarily chosen as in batch mixers, but is determined by operating conditions, screw geometry, and extruder length (residence time). Evaluation of single- and twin-screw extruder designs for various process tasks requires a basic knowledge of key components: gearbox, drive motor, screw shafts, screws, and barrel (Figure 18-4).
Mechanical Description
Gearbox The gearbox performs speed reduction, torque transfer, and thrust loading for the extruder. These functions can be handled separately with individual components or combined in a single unit. Drive motor speed is reduced to operable extruder screw speed in the gearbox. Various gear ratios are available to provide different output speed ranges from standard motor input speeds. Torque is transferred from the drive motor to a single output shaft (single-screw) or must be equally split for dual output
Internal Mixers: Single- and Twin-Screw Extruders
1
2
3 4 5
~ 7
8
91011
12 13
14 15 16
II !
. . . .
7 ,
'
FIGURE 18-4 Extruder components. 1 - T h r u s t bearing assembly; 2 - m o t o r with overload safety clutch; 3 - gear box; 4 - f e e d barrel; 5-electrical resistance heater shells; 6 - kneading elements; 7 - thermocouple for stock temperature; 8 - venting section; 9 - screw elements; 1 0 - stock pressure gauge; 11 - thermocouple for barrel temperature; 1 2 - start-up valve; 13-screen pack changer; 14-pellet/water discharge; 15-UG(under-ground pelletizer); 1 6 - w a t e r inlet. (Source: Werner & Pfleiderer Corporation, Ramsey, NJ.)
shafts (twin-screw). Axial thrust loading (back-pressure) from the screw shaft(s) is taken up by thrust bearings. Single-screw extruders are able to utilize large thrust bearings. Twin-screw extruders, however, must use alternative methods, since the close proximity of the screw shafts precludes the use of large bearings. Twin-screw extruder gearbox designs are inherently more complex than single-screw designs.
Drive Motor Small single-screw (
